JP6529761B2 - Topic providing system and conversation control terminal device - Google Patents

Description

  The present invention relates to a topic providing system for providing a topic to a user, an information search system for providing an information search function to a user, a sentence analysis apparatus for extracting an important character string from text data, and an information update apparatus for comparing a plurality of dictionaries.

  Until now, conversational services such as chat using artificial intelligence have been provided via networks. The conventional conversation control system uses artificial intelligence, and the server analyzes the character information input by the user with the server by morpheme analysis, and generates the answer contents generated in advance that are associated with the semantic contents grasped from the morpheme Output system.

  There is also a conversation control system that divides morphological analysis into first and second morphemes and analyzes according to the concept. Furthermore, there is also a conversation control system that specifies the degree of emotion or the like held by the user and creates an answer.

  In addition, search words that sharply increased in a net search or the like during a predetermined period, words that sharply increased due to news in a predetermined period, and the like are stored as event words, and words that are input together with event words in net search or the like (AND search, etc. The word used in () and the word pre-analyzed from the news including the event word based on the appearance frequency etc. are stored as the event word related word, and in such a situation, 1 of the event word as a search word for net search etc. There is an information retrieval system configured to display a list of event word related words described above when one is input (see Patent Document 4).

  Furthermore, conventionally, there has been disclosed a system for extracting topical words from content input to a computer system and contents such as a home page published on the Internet. For example, in the information acquisition devices disclosed in Patent Document 5 and Patent Document 6, in order to search for information having the most topicality from input information or the like input from the user, the morpheme group stored in the morpheme database and The input information is collated, and the character string corresponding to the morpheme is extracted from the input information. A morpheme corresponds to a minimum unit such as “word” that constitutes a sentence included in input information etc., and this minimum unit includes, for example, a part of speech such as a noun, an adjective, or a verb. In the above morpheme database, morpheme groups including nouns, adjectives, verbs and the like used in general sentences are registered in advance.

  Furthermore, conventionally, a plurality of decomposed text data that can be semantically identified are extracted from text data generated by an individual, and further, filtering is performed using a plurality of conditions. There has been proposed a system for providing space and finding and visualizing specificity.

  For example, Patent Document 7 decomposes personally generated data generated by an individual into a plurality of decomposed text data that can be semantically identified, and predetermined related conditions and predetermined characteristic conditions for objects indicated by the decomposed text data. Data is extracted, a collection of the extracted data is generated as a place of censorship, and further, from the collection of the extracted data, for the target indicated by the decomposed text data, the data satisfying the predetermined specific condition is An information processing system is disclosed that detects and visualizes a predetermined specificity from extracted and data extracted in this way.

  Here, the place of censorship is a space composed of a collection of text data to be censored. Further, the related condition is a condition set by an operator of the information processing system or the like for an object (such as a topic or content included in the text data) indicated by the decomposed text data. For example, the object indicated by the decomposed text data is In the case of food ramen, soy sauce-flavored ramen and salted ramen can be included as related objects that satisfy the relevant conditions.

  The feature condition is a condition regarding an attribute (a feature such as a feature or a feature) of an object indicated by the decomposed text data. The attribute can also be regarded as taste, and for example, when the target indicated by the decomposed text data is food, subjective or objective description by an individual, such as delicious or bad, can be considered.

  The singular condition is to show a predetermined specificity with respect to the subject indicated by the text data. Specificity can be determined depending on various amounts, degrees, and rates of change for a given object, greater or less than a given threshold. For example, if the number of occurrences of the word (speech) “ramen” is greater than that of the remaining partial space in a partial space, the decomposed text data is extracted and visualized as a predetermined specific condition.

JP 2007-305146 A Japanese Patent Laid-Open No. 2004-86246 JP, 2009-245466, A JP 2007-34466 A Patent No. 4253487 gazette Patent No. 4402868 JP, 2012-221222, A

The first problem is as follows.
When proceeding with a conversation, it is common practice to first provide some topic (such as a theme) and proceed with the conversation along the topic. When advancing a conversation with a user using a conventionally existing topic control system (including a conversation control system), it is assumed that the following problems occur in order to adapt to such a conversation form. That is, first of all, even if it is attempted to construct a topic control rule that can respond to various input information in order to realize a natural response that matches the flow of topics and conversations, it is possible to prepare a topic control rule that can respond to all. It can be difficult to achieve a natural response. Furthermore, since the topic control rules are related to advanced and specialized knowledge and techniques throughout the topic control system, separate maintenance operations for changing or modifying the topic control rules are required. Will be difficult.

  The present invention has been made in view of the above-mentioned point, and the purpose of the present invention is to be able to realize a natural response in accordance with the flow of a topic and a conversation, and to change or add a topic control rule. It is an object of the present invention to provide a topic providing system as a topic control system capable of separately performing maintenance work such as correction.

The second issue is as follows.
When proceeding with a conversation, it is common practice to first provide some topic (such as a theme) and proceed with the conversation along the topic. Therefore, in a conventional conversation control system, a system is realized by constructing a huge amount of data and programs for advancing a conversation with the user and performing processing using these data and programs. The Furthermore, in order to be able to respond to various users, most of the processing of the conversation control system was executed on the server. By building such a conversation control system, even with a terminal device (such as a mobile phone or a mobile terminal such as a smartphone) whose storage capacity is relatively small or processing ability is low, the conversation with the user is advanced I was able to go.

  However, in such a conversation control system, it is assumed that the following problems occur. That is, first, in order to execute processing such as a response to input information input by the user on the server side, communication via the network is required each time the user inputs, and the terminal of the user connected to the server When there are many devices, the processing time is long and the response tends to be delayed. Furthermore, in order to execute processing such as a response to input information input from a user on the server side, if different responses can be made according to the status of a plurality of users, the processing time will have to be longer. It was easy to delay the response.

  The present invention has been made in view of the above-described point, and the object of the present invention is to lighten the load on the server and reduce the traffic of the network, and respond appropriately according to the status of a plurality of users. Another object of the present invention is to provide a conversation control terminal device capable of smoothly advancing a conversation in accordance with the flow of a conversation with a user.

The third issue is as follows.
When proceeding with a conversation, it is common practice to first provide some topic (such as a theme) and proceed with the conversation along the topic. Therefore, in a conventional conversation control system, a system is realized by constructing a huge amount of data and programs for advancing a conversation with the user and performing processing using these data and programs. The Therefore, it is possible to provide a service to a plurality of users by connecting the plurality of terminal devices operated by the user to such a server while determining the response information by a server having a large storage capacity and a high processing capacity. It was.

  However, in such a conversation control system, it is assumed that the following problems occur. That is, as described above, in order to conduct a process such as a response to input information from the user to advance a conversation with the user, it is necessary to construct a vast amount of data. The construction of such data required a high level of specialized knowledge and experience throughout the conversation control system. For this reason, in order to make a custom response to each of a plurality of users connected to the server, it has been difficult because it is necessary to work across a wide range of conversation control systems.

  The present invention has been made in view of the above-described point, and the object of the present invention is to be able to make custom responses to each of a plurality of users who use a conversation control system, and to provide An object of the present invention is to provide a maintenance device capable of verifying in advance whether or not a custom is appropriate.

The fourth issue is as follows.
In the information retrieval system as described in Patent Document 4 described above, it is assumed that the following problems occur. That is, when an event word related word is obtained from an AND search with an event word in a net search, such event word related words listed according to the user's input of the event word are derived from the input person Because it is information, it is often biased information that is known to users of the net search, and the association between the event word and the event word related word is also only an association that the input person thinks. Therefore, it is difficult to obtain a new topic that is not known by the event word related word obtained from the event word and the AND search in the net search.

  Also, if the event word related word is pre-analyzed from the news including the event word based on the appearance frequency etc., since the event word related word is an analysis of past news, such an event It is difficult to get the latest topics by word related words.

  An object of the present invention is to provide an information search system, an information search device, an information search method, and a program that can solve the fourth problem described above.

The fifth problem is as follows.
In the information acquisition apparatus disclosed in Patent Document 5 and Patent Document 6 described above, in order to find information (words) having the most topicality from input information etc., the morpheme database to be collated must be prepared in advance. In addition, the effort is extremely large. In the morpheme database, effective analysis of input information and the like can not be realized unless nouns, adjectives, verbs, and the like used in everyday sentences are generally registered.

  Furthermore, in this information acquisition apparatus, topical information can be grasped only in units of morphemes, for example, "words".

  Further, in the above information acquisition apparatus, in order to search for information having the most topicality from input information etc., it is necessary to carry out collation processing with a morpheme database for each word contained in the input information etc. Because it takes a lot of time for access and comparison processing, etc., response time becomes long, and as a result, it is difficult to analyze input information etc. in real time.

  Furthermore, for other languages other than Japanese, in order to find the most topical information from similar processing, ie, input information etc., a morpheme database is newly prepared, and further, according to the grammar of each language , Analysis logic etc. must be rebuilt.

  Therefore, an object of the present invention is to provide a sentence analysis device, a sentence analysis method, and a program that solve all the above-mentioned fifth problems.

The sixth issue is as follows.
The above-described information processing system disclosed in Patent Document 7 extracts a plurality of decomposed text data in which meaning can be identified from text data generated by an individual, and further narrows down decomposed text data satisfying a predetermined condition. Although the text data to be censored contains large amounts of text data that is constantly updated from all over the world on the network, in order to handle the latest information, Extraction must be performed sequentially in line with the update of the enormous text data, and the amount of work is enormous.

  Also, although the newly appeared decomposed text data is of great value as new information, it is extremely difficult to separate the newly appeared decomposed text data from a vast amount of text data. In particular, in such a large amount of text data, extracting such newly appeared decomposed text data is important for understanding the topic, but since the text data is enormous, such decomposed text data is It is difficult to accurately

  Therefore, an object of the present invention is to provide an information updating apparatus, an information updating method, and a program that can solve the sixth problem described above.

The feature according to the first embodiment of the present invention is
An input unit for the user to input input information;
An input information analysis unit that analyzes the input information and generates input specific information;
A scenario data storage unit that extracts scenario data for defining response information on topics;
A response information determination unit that determines the response information based on the scenario data and the input identification information;
And an output unit that outputs the response information determined by the response information determination unit.

  Since the response information is determined based on the scenario data and the input specific information analyzed by the input information analysis unit, it is possible to realize a natural response in accordance with the topic and the flow of the conversation.

  Furthermore, since response information on topics can be defined based on scenario data, changes, additions, modifications, etc. to topic control rules can be made without relying on advanced and specialized knowledge and techniques across topic control systems in general. Maintenance work can be performed separately.

The feature according to the first embodiment of the present invention further includes:
The apparatus further comprises a state control index storage unit that stores a state control index related to the input information and the response information,
The response information determination unit determines the response information by adding the state control index in addition to the scenario data and the input identification information.

  Since the response information is determined using the state control index in addition to the scenario data and the input specific information, it is possible to realize a natural response that is more in line with the topic flow and the flow of conversation.

  Furthermore, the state control index can provide a topic by capturing the history of input and topic provision, so that the user learns context learning (stores past flows of topics and conversations and adapts to the current flow It is possible to realize a response that feels like context learning.

The feature according to the first embodiment of the present invention further includes:
The scenario data includes information defining transitions to different topics,
A topic switching unit that switches scenario data from scenario data for defining response information on the current topic to scenario data for defining response information on the different topic according to the information defining the transition to the different topic. To further provide.

  Since scenario data can be used to switch topics and topic names, it is possible to realize natural responses to the flow of topics and conversations.

  Furthermore, since topic switching can be performed by capturing the user's personality and emotional state by using in combination with the state control index, the user can control emotion (it is called "emotion control" to switch topics by capturing the user's personality and emotional state. It is possible to realize a response that feels

Furthermore, according to the first embodiment of the present invention,
A topic list (e.g., FIG. 17) storing topics each associated with a related word (e.g., a part of text data to which a topic can be associated, which is different from a normal keyword used in a normal search process). The topic list G)
A topic extraction unit (for example, a topic extraction unit that extracts a topic related to the text data from the topic list on the basis of a related term associated with text data (for example, a topic etc. input from a keyboard by a person in charge of the maintenance device 3). A control unit configured by a CPU or the like of the maintenance device 3;
A display control unit (for example, a control unit configured by a CPU or the like of the maintenance device 3) that performs control to display the extracted topic;
The display control means displays, for each of the extracted topics, a related term associated with the relevant topic (for example, as shown in the screen 1815 of FIG. 17, “condition setting: priority related term”) And to display related words in a prioritized list form).

  According to the configuration of the present invention, since the topic itself and a plurality of related words related to the topic can be displayed, the transition from the topic to various topics can be made based on the topic obtained from the topic list. You can display variations.

Furthermore, the features of the first embodiment of the present invention are:
The text data is
(1) Input topic,
(2) It is configured to include at least one of topics extracted based on input topics from external log data (for example, data that can be collected via a network, such as a twitter or blog, etc.) It is.

  According to the configuration of the present invention, since the input of a topic by a direct input device such as a keyboard and the topic obtained from external information are input, these are automatically reduced or added, A topic introduction list can be constructed by repeating.

Furthermore, the features of the first embodiment of the present invention are:
The display control means displays the topic associated with the input related term when the related term associated with the extracted topic is input (for example, as shown in the screen 1815 of FIG. 17). Then, by the input of "reliable", the topic displayed is narrowed down to only the topic associated with the related word "reliable", and as shown on the screen 1811, by the input of "most scary" , To further narrow down and display the displayed topic only to the topic associated with the related word "the most scary".

  According to the configuration of the present invention as described above, it is possible to easily recognize and input the related words associated with the topic, so that the topic displayed as a candidate for the topic introduction list can be a topic introduction list that is more aimed. It can be easily narrowed down.

Furthermore, the features of the first embodiment of the present invention are:
The display control means analyzes an association of related words with reference to a preference dictionary E constructed with other user history in the past, based on an aspect in which related words are associated with a topic in the past, It is to control the display of the topic.

  According to the configuration of the present invention as described above, the aspect relating to the association of related words is analyzed from the response history of another user, so that the related words can be displayed based on the preference of the user.

The features of the second embodiment of the present invention are:
An input unit for the user to input input information;
A scenario data storage unit for storing scenario data for defining response information on topics;
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the state control index;
And an output unit that outputs the response information determined by the response information determination unit.

  Since both the scenario data storage unit and the state control index storage unit are provided in the conversation control terminal device to determine the response information, it is possible to determine and control with the conversation control terminal device whether conversation with the user is possible. The conversation control terminal device does not significantly increase the processing load, and does not increase the load on the server, and also does not increase the network traffic, and smoothly communicates with the flow of the conversation with the user. Can proceed.

  Furthermore, the conversation control terminal device according to the present embodiment determines the response information based on the scenario data and the state control index, so that the response information can be determined according to the progress of the conversation with the user, and the server load is increased. It is possible to respond appropriately according to the state of the user without increasing it.

The feature according to the second embodiment of the present invention further includes:
A transmitter for transmitting the input information to the outside;
A receiving unit that analyzes the transmitted input information and receives generated input identification information;
The response information determination unit determines the response information by adding the input specifying information to the scenario data and the state control index.

  Since the response information is determined by adding the input identification information, the response information can be determined including the input identification information which is the result of analysis by the server or the like outside, and the conversation can be smoothly advanced according to the flow of the conversation with the user Can.

The feature according to the second embodiment of the present invention further includes:
The receiving unit receives scenario data extracted based on the input information;
The scenario data storage unit stores the received scenario data.

  Since scenario data extracted based on input information is received, scenario data can be switched based on input information input by the user, and the conversation can be smoothly advanced according to the flow of conversation with the user it can.

The feature according to the second embodiment of the present invention further includes:
The scenario data includes information defining transitions to different topics,
And a switching input information input unit that generates topic switching input information according to information defining transition to the different topics,
The transmission unit transmits the topic switching input information to the outside.
The receiving unit is to receive scenario data based on the topic switching input information.

  The topic can be switched based on the state control index and the scenario data, so that the topic can be switched while watching the state of the conversation with the user, and the conversation can be smoothly advanced according to the flow of the conversation with the user Can.

The feature of the third embodiment of the present invention is
A scenario data storage unit that stores scenario data for defining response information related to a topic based on input specific information generated by analyzing input information input by a user;
A receiving unit that receives the input identification information;
A scenario data editing unit for making the scenario data editable;
A scenario data verification unit that makes it possible to verify the response of the edited scenario data based on the input identification information received by the reception unit;
And a scenario data transmitting unit for transmitting the edited scenario data to the outside.

  Data in the conversation control system includes both data for analyzing input information input by a user to generate input identification information and scenario data for determining response information based on the input identification information. The scenario data is data that can diversify response information that is a response to the user. According to the configuration described above, the scenario data editing unit can edit the scenario data, and the scenario data verification unit can verify the response of the edited scenario data. In this way, scenario data can be customized for each of a plurality of users who use the conversation control system, without highly specialized knowledge and skills throughout the conversation control system. it can.

The feature according to the third embodiment of the present invention further includes:
An input unit for the user to input input information;
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the state control index;
And an output unit for outputting the response information determined by the response information determination unit; and a terminal device virtual construction unit for virtually constructing a conversation control terminal device.

  The terminal device virtual construction unit can virtually construct and execute the conversation control terminal device in the maintenance device. Therefore, an environment similar to a conversation control terminal used by a general user can be realized in the maintenance device. By this, it is possible to confirm in advance the content and operation of scenario data in an environment similar to the environment in which the user actually conducts a conversation, and to verify the content of the scenario data before having a conversation with the user. It can be verified in advance whether the custom applied to each of a plurality of users connecting to is appropriate.

The feature according to the third embodiment of the present invention further includes:
It further comprises a topic analysis unit for generating a topic list to which the topic approach and connection are given via the related words that relate topics,
The scenario data editing unit can edit, as the scenario data, a topic introduction scenario for introducing a topic to a user using the topic list and the related words, and an input related scenario for responding to user input. It is.

  Data in the conversation control system includes both data for analyzing input information input by a user to generate input identification information and scenario data for determining response information based on the input identification information. The scenario data is data that can diversify response information that is a response to the user. According to the configuration described above, the scenario data editing unit can edit the scenario data in cooperation with the topic analysis unit, and the scenario data verification unit can verify the response of the edited scenario data. In this way, scenario data can be customized for each of a plurality of users who use the conversation control system, without highly specialized knowledge and skills throughout the conversation control system. it can.

The fourth embodiment of the present invention is
From text data collected by keyword-based search (for example, external log 502 (text data) acquired from the collected WEB page shown in FIG. 20), sentence information related to the keyword (for example, text hit in keyword search) Sentence information acquisition means (for example, an input information analysis unit 41 shown in FIG. 20) for acquiring a question sentence or the like which is a part of data;
From each of the sentence information, one or more character strings satisfying a predetermined condition (e.g., a related word which is a character string that can be identified in meaning) is selected, and the character string is stored for each corresponding sentence information Character string selection means (for example, the sentence analysis unit 43 shown in FIG. 20) stored in the means (for example, the related word dictionary 50 shown in FIG. 20);
Information for displaying the selected character string to the user for each corresponding sentence information (for example, display related word / co-occurrence word list display screen 650 as shown in FIG. 37 for the user) 20. An information retrieval system configured to be provided with an information output unit (for example, an input information analysis unit 41 shown in FIG. 20) for outputting input identification information including related words and co-occurrence word data 52 shown in FIG. (For example, the information search system 100 shown in FIG. 20).

  According to such a configuration of the present invention, the selected character string can be obtained as information derived from the information originator, and a new topic that is not known can also be acquired. In addition, since the selected character string (related words) is the latest information obtained from the search based on the keyword, the latest information can be obtained.

The feature of the fourth embodiment of the present invention is
The character string selection means
The present invention is configured to select the character string without performing collation with character string data (for example, a dictionary including morpheme data etc.) stored in advance.

  With such a configuration of the present invention, it is not necessary to create and maintain a dictionary including morpheme data etc. in advance, and an information search system can be easily constructed.

The feature according to the fourth embodiment of the present invention further includes:
The character string selection means further includes
Character string search means (for example, a character string search processing unit 43b shown in FIG. 21) for searching the same character string from the text data;
For the same character string, it is an index indicating how different (for example, there is a variation) the character appearing immediately before the “same character string” searched, which is the difference degree of the previous adjacent character. Of the number of patterns of characters that appear as adjacent characters), and the degree of difference between adjacent characters after that (for example, how different are the characters that appear immediately after the “same character string found” And a difference degree determination unit (for example, a difference degree determination processing unit 43c shown in FIG. 21) which determines an index based on the number of patterns of characters appearing as a subsequent adjacent character).
Specific character string determination means for determining whether the same character string is a specific character string (for example, a related word) based on the degree of difference between the preceding adjacent character and the degree of difference between the subsequent adjacent character For example, a related term determination processing unit 43d) shown in FIG.
The character string selection means is configured to select the character string from the determined specific character string.

  According to such a configuration of the present invention, related words are determined according to the degree of difference between adjacent characters, and it is not necessary to perform comparison processing with a dictionary including morpheme data etc. one by one, processing can be speeded up. It is possible to perform display processing of the lyrics in real time.

The feature according to the fourth embodiment of the present invention further includes:
Scenario data storage means (for example, scenario data 28 and scenario data 55 shown in FIG. 20) for storing scenario data (for example, data consisting of statements as shown in FIG. 14) for defining response information on topics;
Response information determination means (for example, response information determination unit 25 shown in FIG. 20) for determining the response information including the selected character string based on the scenario data;
It is configured to further include response information output means (for example, an output control unit 26 shown in FIG. 20) for outputting the response information determined by the response information determination means.

  According to such a configuration of the present invention, it is possible to diversify the display mode in the case of displaying the selected character string (related words) to the user, and to easily execute and manage the editing of the display.

The feature according to the fourth embodiment of the present invention further includes:
Dictionary comparison means (for example, dictionary comparison processing unit 46c shown in FIG. 22);
When the character string selecting unit stores the character string in the character string storage unit, the character string selecting unit stores the character string in a corresponding dictionary (for example, a related word dictionary 50 shown in FIG. 22) according to the collection condition of the text data. Remember
The dictionary comparison means performs comparison processing for comparing a plurality of the dictionaries, and stores the comparison result in comparison result storage means (for example, comparison result data 54 shown in FIG. 22).
The response information determination means determines the response information including the comparison result based on the scenario data,
The response information output unit outputs the response information determined by the response information determination unit,
The dictionary comparison means is further configured to perform the comparison process when at least one of the plurality of dictionaries is updated, and to automatically update the comparison result stored in the comparison result storage means. It is.

  According to such a configuration of the present invention, it is possible to automatically update the selected character string (related words), and grasp and display the change in the appearance state of the related words according to the comparison result at each update timing. By this, it is possible to diversify the display mode when displaying the related words to the user.

The feature according to the fourth embodiment of the present invention further includes:
The information output means is
When one of the character strings corresponding to one of the sentence information and one of the character strings corresponding to sentence information different from one of the sentence information are in common, the character corresponding to the one sentence information It is configured to output information for associating and displaying a set of columns and the set of character strings corresponding to the other sentence information.

The feature according to the fourth embodiment of the present invention further includes:
The information output means is
Outputting information for displaying all the sets of the character strings corresponding to one or more predetermined sentence information items;
The display order of the character string may be configured to be determined according to a use mode of the user for the character string.

In the fourth embodiment of the present invention, further,
Keyword input means (for example, an input control unit 21 shown in FIG. 20) for inputting a keyword that is a collection condition of text data;
When one or more character strings satisfying a predetermined condition are selected from each of sentence information on the keyword acquired from text data collected based on the keyword, the selected character string is selected. 20. An information retrieval apparatus (eg, FIG. 20) configured to include an information output unit (eg, output control unit 26 shown in FIG. 20) for outputting information to be displayed to the user for each corresponding sentence information. Conversation control terminal device 2 '' shown in FIG.

In the fourth embodiment of the present invention, further,
A sentence information acquisition step of acquiring sentence information related to the keyword from text data collected by a search based on the keyword;
A character string selecting step of selecting one or more character strings satisfying a predetermined condition from each of the sentence information, and storing the character string in the character string storage means for each corresponding sentence information;
The information search method may further include an information output step of outputting information for displaying the selected character string to the user for each corresponding sentence information.

In the fourth embodiment of the present invention, further,
On the computer
Sentence information acquiring means for acquiring sentence information on the keyword from text data collected by a search based on the keyword;
Character string selection means for selecting one or more character strings satisfying a predetermined condition from each of the sentence information, and storing the character string in the character string storage means for each corresponding sentence information,
It is a program for functioning as information output means which outputs the information for displaying the said selected character string to a user for every corresponding above-mentioned sentence information.

The fifth embodiment of the present invention is
Character string search means (for example, a character string search processing unit 43b shown in FIG. 21) for searching for the same character string from text data (for example, the external log 502 which is text data acquired from collected WEB pages);
For the same character string, it is an index indicating how different (for example, there is a variation) the character appearing immediately before the “same character string” searched, which is the difference degree of the previous adjacent character. Of the number of patterns of characters that appear as adjacent characters), and the degree of difference between adjacent characters after that (for example, how different are the characters that appear immediately after the “same character string found” And a difference degree determination unit (for example, a difference degree determination processing unit 43c shown in FIG. 21) which determines an index based on the number of patterns of characters appearing as a subsequent adjacent character).
Specific character string determination means for determining whether the same character string is a specific character string (for example, a related word) based on the degree of difference between the preceding adjacent character and the degree of difference between the subsequent adjacent character For example, it is a sentence analysis device (for example, a sentence analysis device including the sentence analysis unit 43 shown in FIG. 21) configured to include the related word determination processing unit 43d shown in FIG.

  According to such a configuration of the present invention, for example, external information published on the Internet is collected as text data through a network, and the character string search unit searches for the same character string from the acquired text data. After the degree of difference between adjacent characters in the same character string is determined by the degree of difference determining means, the same character string is identified based on the degree of difference in adjacent characters determined by the degree of difference determining means. Since the specific character string determination means can specify whether it is a character string (for example, related words) or not, it is possible to extract a specific character string important for identifying a topic from text data without using a dictionary. it can.

The feature of the fifth embodiment of the present invention is
The specific character string determination means
When a plurality of specific character strings are included in the text data, at least one of the difference degree of the preceding adjacent character, the difference degree of the subsequent adjacent character, the character length, and the appearance frequency with respect to each specific character string , And is configured to rank specific strings in the text data.

  According to this structure, when the text data includes a plurality of specific character strings, the specific character string determination unit determines the degree of difference between the preceding and following adjacent characters with respect to each specific character string, the character length, and Since ranking of specific strings in the text data is performed based on at least one of the appearance frequency, for example, not only extracting specific strings related to a certain topic, but also using parameters from various viewpoints Can be ranked, and analysis of specific strings differentiated according to importance, topicality, etc. can be performed for each of a plurality of specific strings.

The feature according to the fifth embodiment of the present invention is further characterized in that the text data is data collected under a predetermined condition (for example, text data of a WEB page or blog generated by an individual and published on the Internet) , Data such as tweet information),
When the specific character string is determined for each of the different text data, the determined specific character string is grouped by the corresponding text data (for example, specified individually for each text data) A set of strings is formed, and a plurality of text data as a whole are managed as a set of sets of specific strings).

  If configured in this way, for example, text data of web pages and blogs published on the Internet generated by an individual, data such as tweet information are collected as text data, and the collected different texts are collected When the specific character string is determined with respect to data, the specific character string determination unit performs processing of grouping the determined specific character string for each corresponding text data, so that each specific character string is determined. As the strings are grouped, as with the ranking described above, each particular string may be differentiated in the respective group, which may improve the analysis freedom.

In the fifth embodiment of the present invention, further,
Searching for the same string from text data;
Determining, for the same character string, the degree of difference between previous adjacent characters and the degree of difference between subsequent adjacent characters;
Determining whether the same character string is a specific character string based on the degree of difference between the preceding adjacent characters and the degree of difference between the subsequent adjacent characters.
According to this structure, it is possible to provide a method having the same operation and effect as the sentence analysis device of the present invention.

  According to such an embodiment of the present invention, it is possible to extract a specific character string important for identifying a topic from text data without using a dictionary.

In the fifth embodiment of the present invention, further,
On the computer
Character string search means for searching the same character string from text data,
A difference degree determination unit that determines the degree of difference between the preceding adjacent characters and the degree of difference between the subsequent adjacent characters for the same character string;
It is a program that functions as a specific character string determination unit that determines whether the same character string is a specific character string based on the degree of difference between the preceding adjacent characters and the degree of difference between the subsequent adjacent characters.

  According to such a configuration of the present invention, it is possible to extract a specific character string important for identifying a topic from text data without using a dictionary.

The sixth embodiment of the present invention is
A character string (for example, related words) that can be identified in meaning is extracted from text data (for example, external log 502 which is text data acquired from a collected web page), and the extracted character string is used as the text data Character string extraction means (for example, a character string extraction processing unit 46b shown in FIG. 22) stored in a dictionary (for example, a related word dictionary 50) corresponding to
A comparison process of comparing a plurality of the dictionaries (for example, comparing related words stored in related word dictionary (i-1) with related words stored in related word dictionary (i), using related word dictionary (i) A process of determining whether or not there is a newly arrived new arrival related term, or whether or not there is an extinct related term that has disappeared in the related term dictionary (i), etc. is performed, and the comparison result (for example, new arrival related term etc.) Dictionary comparison means (for example, dictionary comparison processing unit 46c shown in FIG. 22) stored in the comparison result storage means (comparison result data 54);
Text data associated with different dictionaries (for example, text data 1 which is an extraction source of related words stored in related word dictionary (i-1), and related words stored in related word dictionary (i) The extraction source text data 2) is collected by different collection conditions (for example, text data at different times searched by the same search condition, or text data searched by different search conditions at the same time) Text data,
The dictionary comparison means performs the comparison process when at least one of the plurality of dictionaries is updated, and automatically updates the comparison result stored in the comparison result storage means (for example, overwrite update, or It is an information updating apparatus (for example, an information updating apparatus including the information updating unit 46 shown in FIG. 22) configured to update cumulatively).

  According to such a configuration of the present invention, for example, a character string that can be semantically identified is extracted from text data, and the extracted character string is stored in a dictionary corresponding to the text data (text data associated with different dictionaries Are text data collected under different collection conditions, and the dictionary comparison means compares a plurality of the dictionaries and automatically stores the comparison results in the comparison result storage means, so that collection is performed under different collection conditions. Character strings that can be semantically identified can be automatically extracted sequentially from the text data, and the latest character strings can be grasped by comparing dictionaries storing those character strings. It is possible to effectively grasp the change of the topic.

The feature of the sixth embodiment of the present invention is
The dictionary comparison means
In the comparison process, when the appearance status of character strings in a plurality of the dictionaries (for example, when the related word dictionary (i-1) and the related word dictionary (i) are compared, the related words are new in the related word dictionary (i) The corresponding character string is stored as the comparison result in accordance with the situation where it appears, the situation where the related term disappears in the related term dictionary (i), etc.).

  According to this configuration, the dictionary comparison unit stores the corresponding character string as the comparison result according to the appearance status of the character string in the plurality of dictionaries in the comparison process, for example, (time series In the following related words dictionary, the related words newly appeared in the related words dictionary, the related words deleted in the later related words dictionary, etc. can be grasped according to the appearance situation.

The feature according to the sixth embodiment of the present invention is further characterized by:
The character string extraction unit
When a plurality of character strings are extracted from the text data, the plurality of character strings are associated and stored in the corresponding dictionary (for example, a plurality of related words extracted from one text data (co-occurrence Related words are ranked and stored as one record),
The dictionary comparison means
In the comparison processing, when there are character strings common to a plurality of the dictionaries, character strings associated with each of the common character strings are compared (for example, commonality is determined between sets of related words) That is, they are configured to compare co-occurrence related words of respective common related words.

  With such a configuration, the character string extraction unit associates, for example, a plurality of character strings and stores them in a dictionary, and the comparison processing is a character string common to a plurality of dictionaries in the comparison processing Since the comparison is also performed with respect to the character strings (vicinity related words) associated with each other, it is possible to grasp the level of commonality or the like regarding the common character string.

In the sixth embodiment of the present invention, further,
A character string extraction step of extracting a character string that can be semantically identified from text data, and storing the extracted character string in a dictionary corresponding to the text data;
Performing a comparison process of comparing a plurality of the dictionaries, and storing a comparison result in comparison result storage means;
The text data associated with the different dictionary is text data collected by different collection conditions,
The dictionary comparison step is configured to perform the comparison process when at least one of the plurality of dictionaries is updated, and to automatically update the comparison result stored in the comparison result storage means. It is an update method.
According to this structure, it is possible to provide an information updating method having the same effects as the information updating device of the present invention.

In the sixth embodiment of the present invention, further,
On the computer
Character string extraction means for extracting a character string capable of semantic identification from text data and storing the extracted character string in a dictionary corresponding to the text data;
A program that performs a comparison process of comparing a plurality of the dictionaries, and causes the comparison result to be stored as comparison result storage means in the dictionary comparison means.
The text data associated with the different dictionary is text data collected by different collection conditions,
Furthermore, the dictionary comparison unit is configured to perform the comparison process when at least one of the plurality of dictionaries is updated, and to automatically update the comparison result stored in the comparison result storage unit. Ru.
According to this structure, it is possible to provide a computer-executable program having the same effects as the information updating apparatus of the present invention.

The effect of the first embodiment of the present invention is
It is possible to realize natural responses to the flow of topics and conversations, and to separately perform maintenance operations such as change, addition, and correction of topic control rules. Further, since the topic itself and a plurality of related words related to the topic can be displayed, it is possible to display variations of various topics by transitioning from the topic based on the topic obtained from the topic list.

The effect of the second embodiment of the present invention is
The load on the server can be lightened, the traffic on the network can be reduced, and it is possible to respond appropriately according to the status of a plurality of users, and to proceed with conversation smoothly according to the flow of conversation with the user.

The effect of the third embodiment of the present invention is
While being able to apply and respond to each of a plurality of users who use a conversation control system, it is possible to verify in advance whether the applied custom is appropriate.

The effect of the fourth embodiment of the present invention is
It is possible to provide a user who uses the information search system with a keyword (character string) capable of acquiring an unknown latest topic.

The effect of the fifth embodiment of the present invention is
With the sentence analysis apparatus according to the present invention, important character strings can be extracted from the text data without using a database to be compared with the text data, whereby a database such as a morpheme database can be created in advance, It is not necessary to maintain and manage. Also, because it does not grasp strings based on morphemes, it is possible to make important strings flexible, for example, a phrase, a clause, or a part of a sentence other than a unit such as a word. Can be grasped in the following manner.

  The same character string is searched from text data by the sentence analysis device according to the present invention, the degree of difference between adjacent characters before and after the character string is determined, and the importance of the character string is determined based on the degree of difference. Therefore, access to the database and comparison processing with morphemes stored in the database are not performed, and important character strings are extracted at high speed from the text data.

  Furthermore, since the importance of the character string is judged by the sentence analysis device according to the present invention on the basis of the degree of difference between adjacent characters before and after the character string as described above, the analysis logic etc. are not reconstructed. It can easily cope with languages other than Japanese.

The effect of the sixth embodiment of the present invention is
The information updating apparatus according to the present invention can automatically extract sequentially resolved text data from text data collected under different collection conditions, and as a result, from the vast amount of text data, the latest resolved text data etc. It is something that can be grasped. Text data may include not only data that is frequently created and updated by individuals on the network, but also data that is created and updated daily at a predetermined institution or the like. In addition, the extraction of decomposed text data is automatically performed, so that it is possible to greatly reduce the effort for defining and creating the decomposed text data.

  Further, the information update device according to the present invention automatically extracts sequentially decomposed text data from text data collected under different collection conditions, and compares the extracted decomposed text data in this way to newly appear. Decomposition text data can be separated, new information with high value can be obtained instantaneously with regard to understanding of topics, etc., and changes in topics can be effectively grasped. In addition, it is possible to grasp the decomposed text data which has disappeared, the decomposed text data which has appeared again, and the like, and the present invention can be applied to various systems which require such information.

  Furthermore, when decomposing text data is extracted from text data by the information updating apparatus according to the present invention, a plurality of decomposing text data extracted from the text data are regarded as a meaningful set, and appearance status of the decomposing text data By associating the above-mentioned set (co-occurrence relation) of the decomposed text data, the change of the topic can be grasped more effectively.

It is a figure which shows the outline | summary of a topic provision system. It is a figure which shows the outline | summary of a conversation control terminal device. It is a figure showing an outline of a maintenance device. FIG. 1 is a block diagram schematically showing a system configuration of a topic providing system 1; It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. It is a flowchart which shows the concrete processing of the statement of scenario data. FIG. 6 is a diagram showing an example output to an output unit 220. It is a figure which shows the concrete example of the statement of scenario data. It is a figure which shows the process of producing | generating response information based on topic analysis, and outputting response information to an output part. It is a figure which shows the example of a structure of a topic introduction list. It is a figure which shows the process of extraction of a topic, the production | generation of a related-words dictionary, and the production | generation of a preference dictionary. The flow of processing from generation of a topic introduction list to output based on the related term dictionary and the topic list when a topic is input to the topic material setting screen of the maintenance device 3 is shown. It is a figure for demonstrating the technical thought of an information retrieval system. It is a figure showing an outline of an information retrieval system. It is a figure which shows the outline | summary of the sentence analysis part of an information retrieval system. It is a figure which shows the outline | summary of the information update part of an information retrieval system. It is a figure which shows the screen transition of a FAQ search system. It is a flowchart showing the display process of a FAQ candidate display screen. It is a figure which shows the outline | summary of the sentence analysis process of an information retrieval system. It is a figure which shows an example of an external log. It is a figure which shows an example of a related-words dictionary. It is a flowchart showing the character string search process of an information retrieval system. It is a figure which shows the structure of the character string search using a suffix array and binary search. It is a figure which shows the structure of the character string search using a suffix array and binary search. It is a flowchart showing the process sequence of the difference degree determination process of an information retrieval system. It is a figure which shows the structure for determining the difference degree of the adjacent character of back and front. It is a flowchart showing the display process of a FAQ display screen. It is a flowchart showing the display process of a related vocabulary and co-occurrence word list screen. It is a figure which shows the example of the screen of a FAQ search system. It is a figure which shows the example of the screen of a FAQ search system. It is a figure which shows the example of the screen of a FAQ search system. It is a figure which shows the example of the preference data, and the example of the screen of a FAQ search system. It is a figure which shows the processing outline of the information update part of an information retrieval system. It is a flowchart showing the process sequence of the character string extraction process of the information update part of an information retrieval system. It is a flowchart showing the process sequence of the dictionary comparison process of the information update part of an information retrieval system. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system with the content of a related-words dictionary. It is a figure which shows the content of the comparison result data memorize | stored by the dictionary comparison process of the information update part of an information retrieval system. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system with the content of a related-words dictionary. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system with the content of a related-words dictionary. It is a figure which shows the character string extraction process and dictionary comparison process of the information update part of an information retrieval system with the content of a related-words dictionary. It is a figure which shows the example of the hardware constitutions of the computer which comprises the subject provision server contained in the information retrieval system of this invention. It is a block diagram which shows the outline of the other system configuration | structure of a topic provision system.

  Hereinafter, the present embodiment will be described based on the drawings.

  FIG. 1 is a diagram showing an overview of the topic providing system 1. FIG. 2 is a diagram showing an outline of the conversation control terminal device 2. FIG. 3 is a diagram showing an outline of the maintenance device 3.

<<< Overview of topic provision system >>>
As shown in FIG. 1, the feature of the topic providing system 1 according to the present embodiment is:
An input unit for the user to input input information;
An input information analysis unit that analyzes the input information and generates input specific information;
A scenario data storage unit that extracts scenario data for defining response information on topics;
A response information determination unit that determines the response information based on the scenario data and the input identification information;
And an output unit that outputs the response information determined by the response information determination unit.

  The topic providing system 1 according to the present embodiment mainly includes an input unit, an input information analysis unit, a scenario data storage unit, a response information determination unit, and an output unit, as shown in FIG. In FIG. 1, these configurations are shown by solid squares. The transmission unit, the reception unit, and the switching input information input unit, which are indicated by dotted squares, will be described later. The portion enclosed by a broken line in FIG. 1 is the configuration of the conversation control terminal device 2 shown in FIG. 2 described later.

  The input unit is a member or part for the user to input input information. The input part should just be what can input the information for which a user desires as input information. For example, the input unit includes a keyboard, a touch panel, a microphone, a camera, and the like. The user can input text data, voice data, image data and the like from the input unit. The input information input to the input unit is supplied to an input information analysis unit described below. The input information is preferably supplied to the input information analysis unit via a transmission unit described later.

  The input information analysis unit analyzes the input information and generates input specific information. The input identification information is information generated as a result of analyzing various types of information included in the input information. For example, there is statistical analysis such as the number and frequency of specific information (such as related words described later) included in input information. Furthermore, analysis of the input information can analyze the intention and preference of the user from the question or the like input by the user. Furthermore, relative analysis results can also be obtained by comparison with other users. Furthermore, data such as an analysis dictionary can be generated in advance, and input information can be analyzed by the analysis dictionary. The input information analysis unit supplies the generated input identification information to a response information determination unit described later.

  The scenario data storage unit is a member or part for extracting scenario data. The scenario data is data stored in advance in a scenario data storage unit (a plurality of scenario data) described later. The scenario data determined to be necessary based on the input identification information generated by the input information analysis unit is extracted, and the extracted scenario data is stored in the scenario data storage unit. The scenario data extracted from the scenario data storage unit (a plurality of scenario data) is stored in the scenario data storage unit via a reception unit described later and a response information determination unit.

  The scenario data is data for defining response information on a topic provided to the user. That is, the scenario data includes topic information to be provided to the user.

  Furthermore, the scenario data also includes a topic introduction list consisting of information on a plurality of topics. The topic introduction list is provided to the user. The user can make an alternative to the input of the input information by selecting a topic included in the provided topic introduction list. Since the user can advance the conversation by the selection operation, the user's input operation can be simplified as compared with the case of inputting characters, and the conversation can be advanced smoothly.

  Also, a plurality of topics can be provided to the user by the topic introduction list, the user can learn various topics, and the range of interest can be expanded.

  Furthermore, the scenario data preferably includes information for greeting the user. By including not only information for providing topics to the user but also information for greetings in the scenario data, it is possible to greet the user and make the conversation more natural.

  The scenario data is data that is created in advance including information that the contractor of the topic providing system 1 wants to provide to the user. The scenario data can define the information to be provided to the user.

  Furthermore, scenario data consists of multiple statements. The statements include output information, output commands and control commands.

  The output information is information output by the output unit. The output information includes topic information, greeting information, and the like. The topic information and the greeting information are information provided to the user for promoting a conversation with the user.

  The output command is a command for controlling the specification of an output when topic information or greeting information is output to the output unit. For example, it is a command for erasing the screen, line feed, controlling the output time, or displaying a predetermined image.

  The control command is a command for determining a statement control, switching a topic name (for example, a theme or the like), or changing a state control index. In particular, the judgment includes judgment for branching according to time and time, judgment for branching according to the contents of the state control index, and the like. By branching by judgment, it is possible to transition to a predetermined statement.

  In the present embodiment, the statements are synonymous with scenario data when it is not necessary to distinguish each of the statements.

  Note that it is not necessary to include all of the output information, the output command, and the control command in a single statement. For example, a predetermined statement can be configured with only output information, can be configured with only output commands, or can be configured with only control commands.

  Thus, the statement includes not only the information to be provided to the user, but also various commands. As a result, in the output unit, topic information and greeting information can be output in various specifications, and conversation can be smoothly and smoothly progressed.

  As described above, the scenario data (statement) includes output information, an output command, a control command, and the like. Topic control rules can be constructed by appropriately defining output information, output commands, and control commands. In particular, by including various information (topics) which the contractor of the topic providing system 1 wants to provide to the user in the scenario data, it is possible to construct a topic control rule desired by the contractor.

  The scenario data may be defined appropriately for the information that the contractor wants to provide, so that changes, additions, corrections, etc. to the topic control rules can be made without relying on advanced and specialized knowledge or technology over the topic provision system 1 in general. Maintenance work can be performed separately.

  The response information determination unit determines response information. The response information is determined based on the scenario data supplied from the input information analysis unit described above and the input identification information. That is, response information is determined using the input identification information obtained by analyzing the input information input by the user. Therefore, the intention of the contractor of the topic providing system 1 can be reflected in the response information by the scenario data, and the intention of the user can be reflected in the response information by the input specific information. It is possible to reflect on it and generate response information. A natural response can be realized by balancing the contractor and the user of the topic providing system 1 and advancing the conversation smoothly.

  The response information includes statements of scenario data. Not only the output information provided to the user included in the statement but also various commands such as output commands can be included in the response information. By doing this, it is possible to output topic information and greeting information from the output unit in various specifications.

  The output unit outputs the response information determined by the response information determination unit. The user is provided with a topic by recognizing the response information output from the output unit.

  Thus, the topic providing system 1 according to the present embodiment can provide various topics to the user by the response information output to the output unit. That is, the topic providing system 1 according to the present embodiment generates input specifying information from input information, determines response information from scenario data and input specifying information, and provides various topics to the user by outputting response information. .

  The topic providing system 1 according to the present embodiment determines response information based on scenario data (for example, a statement and a topic introduction list to be described later, etc.) and input specific information analyzed by the input information analysis unit. It is possible to realize a natural response in line with the flow of

  Furthermore, since response information on topics can be defined based on scenario data, changes, additions, modifications, etc. to topic control rules can be made without relying on advanced and specialized knowledge and techniques across topic control systems in general. Maintenance work can be performed separately.

Furthermore, as shown in FIG. 1, the feature of the topic providing system 1 according to the present embodiment is:
The apparatus further comprises a state control index storage unit that stores a state control index related to the input information and the response information,
The response information determination unit determines the response information by adding the state control index in addition to the scenario data and the input identification information.

  The topic providing system 1 includes a state control index storage unit. The state control index storage unit stores a state control index. The state control index is an index related to input information and response information. The state control index is an index mainly based on the history of conversation, and further is an index indicating the user's emotion or personality that can be determined based on the history of conversation. For example, there are an index determined based on input information input by the user in the past, an index determined based on response information provided to the user in the past, and the like. Furthermore, there are also indicators etc. that indicate the user's emotion, personality, etc. obtained from the user's input to the response information provided to the user in the past.

  The response information determination unit determines response information by adding a state control index in addition to the scenario data and the input identification information. As described above, since the response information is determined using the state control index, providing a topic or advancing the conversation based on the past conversation with the user, the user's emotion or character obtained from the conversation, etc. Can. Therefore, it is possible to prevent the same topic from being provided redundantly to the user or to provide the leaped topic to the user, and it is possible to advance a smooth conversation in accordance with the user's emotion or personality.

  In the present embodiment, the state control index storage unit is provided in the conversation control terminal device 2. The state control indicator is preferably determined or changed by the response information determination unit. The state control index is preferably determined by the response information determination unit based on the scenario data and the input identification information.

  The topic providing system 1 according to the present embodiment uses the state control index in addition to the scenario data and the input specific information to determine the response information, so that a natural response that more closely matches the flow of the topic and the flow of conversation can be realized. Can.

Furthermore, as shown in FIG. 1, the feature of the topic providing system 1 according to the present embodiment is:
The scenario data includes information (for example, topic switching information to be described later) defining transition to a different topic,
A topic switching unit that switches scenario data from scenario data for defining response information on the current topic to scenario data for defining response information on the different topic according to the information defining the transition to the different topic. To further provide.

  In the present embodiment, the topic switching unit includes a switching input information input unit and an input information analysis unit. The topic switching unit determines, based on the scenario data and the input identification information, whether to switch the scenario data. Specifically, it is preferable to switch scenario data based on the state control index. In the switching input information input unit, the topic switching input information is transmitted to the transmission unit described later, and is supplied to the input information analysis unit. The input information analysis unit extracts scenario data stored in the scenario data storage unit (a plurality of scenario data) based on the topic switching input information as well as the input information.

  The scenario data includes information defining transition to different topics (topic names). The topic switching unit switches scenario data. This switching is to switch from scenario data for defining response information on the current topic to scenario data for defining response information on a different topic.

  For example, all scenario data is stored in a server or the like, scenario data reassembled from all scenario data of the server is generated based on information defining transition to different topics, and scenario data reorganized is It is stored in the scenario data storage unit. This rearrangement of scenario data can be performed by determining a combination of a plurality of statements according to the topic name.

  Therefore, even when the user's conversation with the user makes a transition from one topic to another, transition to another topic can be made according to the information defining transition to a different topic, and all topics can be handled, It can correspond to various users. Other topics can be provided to the user by the rearranged scenario data.

  Also, since scenario data may be prepared for each topic (topic name), maintenance of scenario data is facilitated. Specifically, when the scenario data needs to be changed, it is sufficient to correct only the scenario data. Also, when a new topic is needed, it is sufficient to add only that scenario data. Furthermore, when the old topic is no longer needed, only the scenario data may be deleted.

  The scenario data may be defined appropriately for each topic (topic name) about the information that the contractor wants to provide, so even if the number of topics increases, advanced and specialized knowledge over the entire topic providing system 1 Maintenance tasks such as changing, adding, and modifying topic control rules can be performed on a topic-by-topic basis, without relying on technology and technology.

  Since the information defining the transition to a different topic is included, even when such scenario data is updated, it is possible to easily match the transition of the scenario data.

  For example, it is preferable to determine whether or not to switch topics based on a character index, which will be described later, among the state control indicators. The character index is an index indicating whether the user is positive or negative with respect to the topic.

  Since the topic providing system 1 according to the present embodiment can switch topics using scenario data, it is possible to realize a natural response in accordance with the flow of topics and conversations.

  The transmission unit, the reception unit, and the switching input information input unit, which are shown by the dotted square in FIG. 1, will be described. The topic providing system 1 according to the present embodiment can include the transmission unit, the reception unit, and the switching input information input unit.

  Further, the transmission unit shown in FIG. 1 is an apparatus or a member for transmitting input information to the outside. The input information may be transmitted to the outside. The outside can be, for example, a server, the conversation control terminal 2 or the like.

  Furthermore, the receiving unit shown in FIG. 1 is an apparatus or a member for receiving input identification information. The input specific information is generated externally. That is, the receiving unit is an apparatus or a member that receives externally generated input identification information. The outside analyzes the input information transmitted from the transmission unit to generate input identification information, and the generated input identification information is transmitted to the reception unit.

  Furthermore, the switching input information input unit illustrated in FIG. 1 generates topic switching input information (for example, a character index or the like to be described later) according to the information that defines the transition to a different topic. The information defining the transition to a different topic is the information included in the scenario data, for example, topic switching information described later.

  Preferably, the topic switching unit described above includes an input information analysis unit and a switching input information input unit. Since the topic switching input information is generated based on the input information, it is possible to make transition to a topic that reflects the user's intention.

  The scenario data storage unit (a plurality of scenario data) illustrated in FIG. 1 stores a plurality of scenario data. Here, the plurality of scenario data are all scenario data corresponding to a topic name required to have a conversation with the user. Among all scenario data, scenario data determined to be necessary based on input specific information is extracted.

<<< Overview of Conversation Control Terminal Device 2 >>>
As shown in FIG. 2, the features of the conversation control terminal device 2 according to the present embodiment are:
An input unit for the user to input input information;
A scenario data storage unit for storing scenario data for defining response information on topics;
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the state control index;
And an output unit that outputs the response information determined by the response information determination unit.

  As shown in FIG. 2, the conversation control terminal device 2 according to the present embodiment mainly includes an input unit, a scenario data storage unit, a state control index storage unit, a response information determination unit, and an output unit. In FIG. 2, these configurations are indicated by solid squares. The portion surrounded by a large broken line in FIG. 2 is the configuration of the conversation control terminal device 2. The scenario data storage unit (a plurality of scenario data) and the input information analysis unit, which are indicated by dotted line squares, are included in the topic providing server 4.

  The input unit is, like the input unit of the topic providing system 1 according to the present embodiment, a member or a part for the user to input input information. The input part should just be what can input the information for which a user desires as input information. For example, the input unit includes a keyboard, a touch panel, a microphone, a camera, and the like. The user can input text data, voice data, image data and the like from the input unit. The input information input to the input unit is supplied to an input information analysis unit described below. The input information is preferably supplied to the input information analysis unit via a transmission unit described later.

  The scenario data storage unit is a member or part for storing scenario data. The scenario data is data stored in advance in a scenario data storage unit (a plurality of scenario data) of the topic providing server 4 shown in FIG. The scenario data determined to be necessary based on the input identification information generated by the input information analysis unit is extracted, and the extracted scenario data is stored in the scenario data storage unit. The scenario data extracted from the scenario data storage unit (a plurality of scenario data) is stored in the scenario data storage unit via a reception unit described later and a response information determination unit.

  The scenario data is data for defining response information on a topic provided to the user. The scenario data includes topic information to be provided to the user. The configuration, function, etc. of scenario data used in the conversation control terminal device 2 are the same as the scenario data of the topic providing system 1 according to the present embodiment.

  The conversation control terminal device 2 according to the present embodiment includes a state control index storage unit. The state control index storage unit stores a state control index. The state control index is an index mainly based on the history of conversation, and further is an index that indicates the user's emotion or personality that can be derived based on the history of conversation. For example, there are an index determined based on input information input by the user in the past, an index determined based on response information provided to the user in the past, and the like. Furthermore, there are also indicators etc. that also indicate the user's emotions and personality that can be derived from the user's input to the response information provided to the user in the past.

  As described above, in the present embodiment, the conversation control terminal device 2 includes the state control index storage unit. That is, the configuration is not such that the topic providing server 4 or the like has a state control index storage unit outside the conversation control terminal device 2. Therefore, in the present embodiment, the external device such as the topic providing server 4 does not control the conversation with the user, but the conversation control terminal device 2 controls the conversation with the user.

  The response information determination unit determines response information based on the scenario data and the state control index supplied from the input information analysis unit of the topic providing server 4. The scenario data enables the contractor's intention of the topic providing system 1 to be reflected in the response information.

  The response information includes statements of scenario data. Not only the output information provided to the user included in the statement but also various commands such as output commands can be included in the response information. By doing this, it is possible to output topic information and greeting information from the output unit in various specifications.

  Furthermore, since the response information is determined using the state control index, it is possible to provide a topic or to advance the conversation based on the past conversation with the user, the user's emotion or personality obtained from the conversation, etc. . Therefore, it is possible to prevent the same topic from being provided redundantly to the user or providing a leaping topic to the user, and a natural response can be realized by advancing a smoother conversation.

  The state control indicator is preferably determined or changed by the response information determination unit. The state control index is preferably determined by the response information determination unit based on the scenario data and the input identification information.

  The output unit outputs the response information determined by the response information determination unit. The user is provided with a topic by recognizing the response information output from the output unit.

  Since the conversation control terminal device 2 according to the present embodiment is provided with both the scenario data storage unit and the state control index storage unit in the conversation control terminal device 2 to determine the response information, is conversation with the user possible? Whether or not it can be determined by the conversation control terminal device 2 and controlled, and without significantly increasing the processing load on the conversation control terminal device 2 and without increasing the load on the server, the traffic of the network is also increased. The conversation can proceed smoothly according to the flow of the conversation with the user.

  Furthermore, since the conversation control terminal device 2 according to the present embodiment determines the response information based on the scenario data and the state control index, the response information can be determined according to the progress of the conversation with the user, and the server load According to the state of the user without increasing the

Furthermore, as shown in FIG. 2, the features of the conversation control terminal device 2 according to the present embodiment are:
A transmitter for transmitting the input information to the outside;
A receiving unit that analyzes the transmitted input information and receives generated input identification information;
The response information determination unit determines the response information by adding the input specifying information to the scenario data and the state control index.

  Conversation control terminal device 2 according to the present embodiment further includes a transmitter and a receiver.

  The transmitter transmits the input information to the outside. The input information may be transmitted to the outside of the conversation control terminal device 2. The outside can be, for example, a server, another conversation control terminal device 2 or the like.

  The receiving unit receives the input identification information. The input identification information is generated outside the conversation control terminal device 2. That is, the receiving unit is a device or a member that receives the input identification information generated outside the conversation control terminal device 2. Outside the conversation control terminal device 2, the input information transmitted from the transmission unit is analyzed to generate input identification information, and the generated input identification information is transmitted to the reception unit of the conversation control terminal device 2.

  The response information determination unit determines response information by adding the input identification information to the scenario data and the state control index. The response information is determined including the input specific information obtained by analyzing the input information input by the user. Therefore, the intention of the contractor of the topic providing system 1 can be reflected in the response information by the scenario data, and the intention of the user can be reflected in the response information by the input specific information. It is possible to reflect on it and generate response information. A natural response can be realized by balancing the contractor and the user of the topic providing system 1 and advancing the conversation smoothly.

  Since the conversation control terminal device 2 according to the present embodiment adds the input specific information to determine the response information, it can determine the response information including the input specific information which is the result of analysis by the server or the like outside, Conversation can proceed smoothly according to the flow of conversation.

Furthermore, as shown in FIG. 2, the features of the conversation control terminal device 2 according to the present embodiment are:
The receiving unit receives scenario data extracted based on the input information;
The scenario data storage unit stores the received scenario data.

  The receiving unit of the conversation control terminal device 2 according to the present embodiment receives scenario data extracted based on input information. That is, scenario data is extracted outside the conversation control terminal device 2. Based on the input information transmitted to the outside of the conversation control terminal device 2, scenario data is extracted on the basis of the input information outside the conversation control terminal device 2. The extracted scenario data is transmitted to the reception unit of the conversation control terminal device 2.

  The scenario data storage unit stores scenario data received by the reception unit of the conversation control terminal device 2. As described above, the configuration, function, and the like of scenario data used in the conversation control terminal device 2 are the same as the scenario data of the topic providing system 1 according to the present embodiment.

  Since the conversation control terminal device 2 according to the present embodiment receives the scenario data extracted based on the input information, the scenario data can be switched based on the input information input by the user, and the conversation with the user is performed. The conversation can proceed smoothly according to the flow of

Furthermore, as shown in FIG. 2, the features of the conversation control terminal device 2 according to the present embodiment are:
The scenario data includes information (for example, topic switching information to be described later) defining transition to a different topic,
A switching input information input unit configured to generate topic switching input information (for example, a character index to be described later, etc.) according to the information defining the transition to the different topics;
The transmission unit transmits the topic switching input information to the outside.
The receiving unit is to receive scenario data based on the topic switching input information.

  The scenario data used in the conversation control terminal device 2 according to the present embodiment includes information defining the transition to different topics.

  Conversation control terminal device 2 according to the present embodiment includes a switching input information input unit. The switching input information input unit generates topic switching input information in accordance with information defining transition to a different topic. The information defining the transition to different topics is, for example, topic switching information which is an element of a statement described later. Further, the topic switching input information includes, for example, a character index described later.

  The switch input information input unit determines whether to switch scenario data based on the scenario data and the input identification information. Specifically, it is preferable to switch scenario data based on the state control index. In the switching input information input unit, the topic switching input information is transmitted to the transmitting unit, and is supplied to the input information analysis unit. The input information analysis unit extracts scenario data stored in the scenario data storage unit (a plurality of scenario data) of the topic providing server 4 based on the topic switching input information as well as the input information.

  Furthermore, the transmission unit of the conversation control terminal device 2 according to the present embodiment transmits topic switching input information to the outside. Further, the receiving unit receives scenario data based on the topic switching input information.

  The scenario data for defining response information on the current topic can be switched to scenario data for defining response information on different topics.

  For example, all scenario data is stored in a server or the like, scenario data reassembled from all scenario data of the server is generated based on topic switching input information, and scenario data reclassified is the conversation control terminal device 2 Are stored in the scenario data storage unit of

  Therefore, even when the user's conversation with the user makes a transition from one topic to another, transition to another topic can be made according to the information defining transition to a different topic, and all topics can be handled, It can correspond to various users.

  Also, since scenario data may be prepared for each topic (topic name), maintenance of scenario data is facilitated. Specifically, when the scenario data needs to be changed, it is sufficient to correct only the scenario data. Also, when a new topic is needed, it is sufficient to add only that scenario data. Furthermore, when the old topic is no longer needed, only the scenario data may be deleted.

  Of the state control indicators, it is preferable to determine whether to switch topics based on a character indicator to be described later. The character index is an index indicating whether the user is positive or negative with respect to the topic.

  According to the conversation control terminal device 2 according to the present embodiment, since the topic can be switched based on the state control index and the scenario data, the topic can be switched while watching the state of the conversation with the user. The conversation can proceed smoothly according to the flow of the conversation with.

  The input information analysis unit illustrated in FIG. 2 analyzes the input information and generates input identification information. The input identification information is information generated as a result of analyzing various types of information included in the input information. For example, there is statistical analysis such as the number and frequency of specific information (such as related words described later) included in input information.

  The scenario data storage unit (a plurality of scenario data) illustrated in FIG. 2 stores a plurality of scenario data. Here, the plurality of scenario data are all scenario data corresponding to a topic name required to have a conversation with the user. Among all scenario data, scenario data determined to be necessary based on input specific information is extracted.

<<< Overview of Maintenance Device 3 >>>
As shown in FIG. 3, the feature of the maintenance device 3 according to the present embodiment is
A scenario data storage unit that stores scenario data for defining response information related to a topic based on input specific information generated by analyzing input information input by a user;
A receiving unit that receives the input identification information;
A scenario data editing unit for making the scenario data editable;
A scenario data verification unit that makes it possible to verify the response of the edited scenario data based on the input identification information received by the reception unit;
And a scenario data transmitting unit for transmitting the edited scenario data to the outside.

  As shown in FIG. 3, the maintenance device 3 according to the present embodiment mainly includes a scenario data storage unit, a reception unit, a scenario data editing unit, a scenario data verification unit, and a scenario data transmission unit. In FIG. 3, these configurations are indicated by solid squares. The scenario data storage unit (a plurality of scenario data) and the input information analysis unit indicated by dotted line squares are included in the topic providing server 4 (see FIG. 2). As shown in FIG. 3, a scenario data verification unit to which a state control index storage unit is added constitutes a terminal device virtual construction unit to be described later. The information exchanged between these configurations is similar to that of the topic providing system 1 and the conversation control terminal device 2 described above.

  The topic providing system 1 and the conversation control terminal device 2 described above are mainly for a general user to have a conversation with the conversation control terminal device 2. On the other hand, the maintenance device 3 is mainly used by a contractor of the topic providing system 1, and the contractor of the topic providing system 1 performs maintenance of scenario data for providing topics to general users. A device for Although the maintenance device 3 has such a difference, in FIG. 3, the same names are given to configurations having the same functions as the topic providing system 1 and the conversation control terminal device 2 and using the same data.

  The scenario data storage unit and the reception unit in the maintenance device 3 are substantially the same as the scenario data storage unit and the reception unit in the topic providing system 1 and the conversation control terminal device 2 described above. It can be the same as the topic providing system 1 and the conversation control terminal device 2. For example, the scenario data storage unit and the reception unit of the maintenance device 3 may be implemented in the conversation control terminal device 2 assumed to be used by the user. Furthermore, as described later, the conversation control terminal device 2 may be virtually constructed in the maintenance device 3 and may be a scenario data storage unit and a reception unit of the virtual conversation control terminal device 2.

  The scenario data storage unit is a member or part for storing scenario data. The scenario data is data stored in advance in a scenario data storage unit (a plurality of scenario data) of the topic providing server 4 (see FIG. 2). The scenario data determined to be necessary based on the input identification information generated by the input information analysis unit is extracted, and the extracted scenario data is stored in the scenario data storage unit. The scenario data extracted from the scenario data storage unit (a plurality of scenario data) is stored in the scenario data storage unit via a reception unit described later and a response information determination unit.

  The scenario data is data for defining response information related to a topic based on input specific information. The input specific information is information generated by analyzing the input information. The input information is, for example, information input by the user in the conversation control terminal device 2.

  As described above, the maintenance device 3 is mainly used by a subscriber of the topic providing system 1. In the maintenance device 3, the input information can be virtually input by the user. The maintenance device 3 is for verifying scenario data before making the scenario data available to the user. Therefore, the user here may be a virtual user, and information assumed to be input by the actual user may be input information. Therefore, input specification information can be generated using various input information assumed, and the response of scenario data can be verified by the scenario data verification unit described later.

  The scenario data includes topic information to be provided to the user. The configuration, functions, and the like of scenario data used in the maintenance device 3 are the same as the scenario data of the topic providing system 1 according to the present embodiment described above and the scenario data of the conversation control terminal device 2.

  Although the configuration and functions of scenario data are the same as scenario data in topic providing system 1 and conversation control terminal device 2, as described above, maintenance device 3 can use scenario data for the user. The scenario data is to be verified before making Therefore, the scenario data targeted by the maintenance device 3 is scenario data to be verified, and is data before making the scenario data available to the user.

  The receiving unit receives the input identification information. The input specific information is generated outside the maintenance device 3. That is, the receiving unit is an apparatus or a member that receives the input identification information generated outside the maintenance device 3. The input information is analyzed outside the maintenance device 3 to generate input identification information, and the input identification information generated outside the maintenance device 3 is transmitted to the reception unit of the maintenance device 3.

  As described above, the input information can be virtually input by the user. Therefore, the input identification information here can be generated outside the maintenance device 3 as the input information assumed to be input by the actual user. Thus, scenario data can be verified based on various input information by using information assumed to be input by the actual user as the input information.

  The scenario data editing unit is an apparatus or member for making scenario data editable. The scenario data can be edited by the person in charge of the contractor of the topic providing system 1 operating a keyboard or the like. Editing includes adding, deleting, and changing scenario data. Specifically, editing is a process of adding, deleting, and changing statements that constitute scenario data. By editing scenario data, it is possible to customize scenario data for each of a plurality of users.

  The scenario data verification unit is an apparatus or member for making it possible to verify the response of scenario data. The scenario data here is edited scenario data. It is an apparatus or member for verifying whether the response of the scenario data edited by the scenario data editing unit is appropriate.

  As described above, the scenario data verification unit can generate input identification information using various input information assumed, and can verify the response of the scenario data edited by the scenario data editing unit. Therefore, since it is possible to verify whether or not the response of the scenario data is appropriate for every user, the scenario data editing unit can verify the custom of the scenario data applied to each of the plurality of users.

  The scenario data transmission unit transmits the edited scenario data to the outside. The outside can be, for example, a server, another conversation control terminal device 2 or the like. As described above, by transmitting edited scenario data to the outside, verified scenario data can be made available to the user.

  As described above, the scenario data editing unit edits scenario data using the information virtually input by the user as input information, and the scenario data verification unit verifies scenario data as described later. Thus, the topic analysis unit can generate a topic list, and edit and verify scenario data based on the topic list. The topic analysis unit will be described later.

  The scenario data verification unit verifies whether the response of the scenario data edited by the scenario data editing unit is appropriate. By doing this, it is possible to confirm the content and consistency of the scenario data before transmitting the scenario data to the outside of the server or the like. Finally, at least a part of the verified scenario data transmitted to the outside such as a server is transmitted to the speech control terminal device 2 and used for the conversation with the user.

  Data in the conversation control system includes both data for analyzing input information input by a user to generate input identification information and scenario data for determining response information based on the input identification information. The scenario data is data that can diversify response information that is a response to the user. According to the configuration described above, the scenario data editing unit can edit the scenario data, and the scenario data verification unit can verify the response of the edited scenario data. In this way, scenario data can be customized for each of a plurality of users who use the conversation control system, without highly specialized knowledge and skills throughout the conversation control system. it can.

Furthermore, as shown in FIG. 3, the feature of the maintenance device 3 according to the present embodiment is
An input unit for the user to input input information;
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the state control index;
And an output unit for outputting the response information determined by the response information determination unit; and a terminal device virtual construction unit for virtually constructing a conversation control terminal device.

  The maintenance device 3 according to the present embodiment has a terminal device virtual construction unit as shown in FIG. This terminal device virtual construction unit virtually constructs a conversation control terminal device 2 '.

  Furthermore, the maintenance device 3 according to the present embodiment has a scenario data verification unit, as shown in FIG. What added a state control index storage unit to the scenario data verification unit constitutes a terminal device virtual construction unit.

  Thus, the maintenance device 3 is provided with the conversation control terminal device 2 'as a function. The maintenance device 3 can be provided with the function of the conversation control terminal device 2 'as a package for simulation. Alternatively, the conversation control terminal device 2 ′ virtually constructed without the maintenance function can be used as the actual conversation control terminal device 2. The conversation control terminal device 2 'virtually constructed may be realized as hardware or software. When realized as hardware, a separate device different from the actual conversation control terminal device 2 may be a conversation control terminal device 2 'virtually constructed. Further, in the case of software implementation, the conversation control terminal device 2 ′ may be implemented by emulation or the like in the maintenance device 3.

  The maintenance device 3 is mainly used by a contractor of the topic providing system 1, and may be a device for maintaining scenario data for providing topics to general users.

  As shown in FIG. 3, the terminal device virtual construction unit and the scenario data verification unit mainly include an input unit, a response information determination unit, and an output unit. Furthermore, the terminal device virtual construction unit includes a state control index storage unit.

  The input unit is a member or part for the user to input input information. The input part should just be what can input the information for which a user desires as input information. For example, the input unit includes a keyboard, a touch panel, a microphone, a camera, and the like. The user can input text data, voice data, image data and the like from the input unit.

  As described above, here, the input information may be information virtually input by the user. The user may also be a virtual user, and information assumed to be input by the actual user may be input information.

  The state control index storage unit stores a state control index. The state control index is an index related to input information and response information. The state control indicator is mainly an indicator related to the history. For example, there are an index related to input information input by the user in the past and an index related to response information provided to the user in the past.

  The response information determination unit determines response information by adding a state control index in addition to the scenario data and the input identification information.

  The output unit outputs the response information determined by the response information determination unit.

  In this way, in the virtually constructed conversation control terminal 2 ', edited scenario data can be used and verified scenario data can be used, and it becomes possible to use it for the user Before, you can check the content and control of the scenario data. It is possible to provide the user with appropriate topics.

  The terminal device virtual construction unit can virtually construct and execute the conversation control terminal device 2 ′ in the maintenance device 3. Therefore, the maintenance device 3 can realize the same environment as the conversation control terminal device 2 used by a general user. By this, it is possible to confirm in advance the content and operation of scenario data in an environment similar to the environment in which the user actually conducts a conversation, and to verify the content of the scenario data before having a conversation with the user. It can be verified in advance whether the custom applied to each of a plurality of users connecting to is appropriate.

Furthermore, as shown in FIG. 3, the feature of the maintenance device 3 according to the present embodiment is
It further comprises a topic analysis unit for generating a topic list to which the topic approach and connection are given via the related words that relate topics,
The scenario data editing unit can edit, as the scenario data, a topic introduction scenario for introducing a topic to a user using the topic list and the related words, and an input related scenario for responding to user input. It is.

  The maintenance device 3 according to the present embodiment includes a topic analysis unit (not shown). The topic analysis unit is an apparatus or member for generating a topic list. The topic list is data in which the approach and connection of topics are given through the related words that relate topics. The topic analysis unit accumulates related words associated with the topic in the topic list. In the maintenance device 3, the topic list is data provided to the contractor of the topic providing system 1, and is used when generating scenario data.

  For example, a topic list is output to the output unit of the maintenance device 3 as shown in FIG. 17 described later. The contractor of the topic providing system 1 can construct topic introduction scenario data and input related scenario data to be provided to the user with reference to the topic list output to the output unit of the maintenance device 3. By doing this, the contractor of the topic providing system 1 can easily and easily construct data of the topic introduction scenario and data of the input related scenario.

  The scenario data editing unit can edit the topic introduction scenario and the input related scenario as scenario data. Topic introduction scenarios and input related scenarios are made editable by topic lists and related words. Since it can be edited using the topic list and related words, input related scenarios can be constructed easily and easily.

  The topic introduction scenario is a scenario for introducing a topic to the user. The user is provided with a topic by a topic introduction scenario. The input related scenario is a scenario for responding to user's input. When the user inputs predetermined information, for example, information such as a greeting, the input related scenario causes the information such as a corresponding greeting to be replied to the user.

  By the scenario data editing unit, the contractor of the topic providing system 1 can edit the topic introduction scenario and the input related scenario so as to be desired.

  The topic list can be updated based on various logs, eg, obtainable data such as Twitter and blog. That is, the latest information can be accumulated as related words in the topic list. Therefore, when the topic introduction scenario and the input related scenario are edited or verified based on the topic list, the contractor of the topic providing system 1 obtains the latest information by the topic list, and the topic introduction scenario The input related scenario can be edited, and the user can be provided with a topic including the latest information, and the latest information can also open up a new user group.

  Data in the conversation control system includes both data for analyzing input information input by a user to generate input identification information and scenario data for determining response information based on the input identification information. The scenario data is data that can diversify response information that is a response to the user. According to the configuration described above, the scenario data editing unit can edit the scenario data in cooperation with the topic analysis unit, and the scenario data verification unit can verify the response of the edited scenario data. In this way, scenario data can be customized for each of a plurality of users who use the conversation control system, without highly specialized knowledge and skills throughout the conversation control system. it can.

  The transmission unit illustrated in FIG. 3 is an apparatus or a member for transmitting input information to the outside. The input information may be transmitted to the outside. The outside can be, for example, the topic providing server 4 (see FIG. 2), the conversation control terminal device 2 or the like.

  Furthermore, the input information analysis unit shown in FIG. 3 analyzes the input information and generates input identification information. The input identification information is information generated as a result of analyzing various types of information included in the input information. For example, there is statistical analysis such as the number and frequency of specific information (such as related words described later) included in input information.

  Furthermore, the scenario data storage unit (a plurality of scenario data) illustrated in FIG. 3 stores a plurality of scenario data. Here, the plurality of scenario data are all scenario data corresponding to a topic name required to have a conversation with the user. Among all scenario data, scenario data determined to be necessary based on input specific information is extracted.

<<< System Configuration of Topic Providing System 1 >>>
FIG. 4 is a block diagram schematically showing the system configuration of the topic providing system 1.

  The topic providing system 1 includes a topiclet 20, an iWA 30, and an iWA Manager 40.

<< Topiclet 20 >>
In the present embodiment, the Topic 20 corresponds to, for example, hardware such as a terminal device used by a user. Also, Topic 20 corresponds to the conversation control terminal device 2 shown in FIG. The topic is provided to the user by Topic 20. In the present embodiment, “Topiclet” may be used synonymously with software executed by a terminal device to provide a topic to a user, or a topic providing environment that can be realized by the terminal device or the software.

  Specifically, the Topic 20 includes a CPU (central processing unit), a ROM (read only memory), a RAM (random access memory), a display, a keyboard (all not shown), and the like. The Topic 20 can be a personal computer, a portable terminal device, or the like.

  The Topic 20 includes an input unit 210, an output unit 220, a transmission unit 230, a reception unit 240, a response information determination unit 250, a switching input information input unit 260, a scenario data storage unit 270, and a state control index storage unit 280.

<Input unit 210>
The input unit 210 is a device or a member for the user to input input information. The input unit 210 includes a keyboard, a touch panel, a microphone, and the like. The input unit 210 may be any device or member that allows the user to input information such as a question.

<Output unit 220>
The output unit 220 outputs the response information determined by the response information determination unit described later. The output unit 220 includes a display, a speaker, and the like. The output part 220 should just be what can output response information so that a user can recognize.

  Thus, the user can advance the conversation by inputting the input information to the input unit 210 and recognizing the response information output to the output unit 220.

<Transmission Unit 230 and Reception Unit 240>
The transmission unit 230 is a device or a member for transmitting the input information input to the input unit 210 to the iWA 30. For example, the transmission unit 230 includes a communication interface.

  The input information transmitted to the iWA 30 includes a user ID for identifying the user in addition to the input information from the input unit 210. The user ID may be any information that can identify the user. It may be information assigned to each of the users. Furthermore, the user ID may be information that can uniquely identify Topict 20, such as the serial number of Topict 20. The user ID may be any information that can identify each of the users who use the topic providing system 1 and the conversation control terminal device 2.

  The receiving unit 240 is a device or a member for receiving the input identification information and the scenario data transmitted from the iWA 30. For example, the transmission unit 230 includes a communication interface. The Topic 20 is communicably connected to the iWA 30 by the transmission unit 230 and the reception unit 240.

<Response Information Determination Unit 250>
The response information determination unit 250 determines response information based on the input identification information and the scenario data. For example, the response information determination unit 250 includes a CPU of the Topic 20, a ROM, a RAM, and the like. Thus, response information is determined using the input identification information and scenario data transmitted from the iWA 30.

  The response information determining unit 250 dynamically changes the response information based on the scenario data regardless of the input.

  The response information determination unit 250 determines response information. Response information is determined based on scenario data and input identification information. That is, response information is determined using the input identification information obtained by analyzing the input information input by the user. Therefore, response information reflecting the user's intention can be generated, and the conversation with the user can be smoothly advanced by providing a topic desired by the user.

  The response information includes statements of scenario data based on the input specific information. Not only the output information provided to the user included in the statement but also various commands such as output commands can be included in the response information. In this way, the output unit can output topic information and greeting information in various specifications.

  The response information determination unit 250 determines the response information by adding a state control index in addition to the scenario data and the input identification information. As described above, by determining the response information using the state control index, it is possible to provide a topic or advance the conversation based on the past conversation with the user. Therefore, it is possible to prevent the same topic from being provided redundantly to the user or providing a leaping topic to the user, and a smoother conversation can be advanced.

  Further, the response information determination unit 250 may determine the response information based on the scenario data and the state control index.

<Switching input information input unit 260>
The switching input information input unit 260 includes a CPU of the Topic 20, a ROM, a RAM, and the like.

  The switching input information input unit 260 generates topic switching input information in accordance with information defining transition to a different topic. The information defining the transition to a different topic is, for example, topic switching information to be described later. Further, the topic switching input information includes, for example, a character index described later.

<Scenario data storage unit 270>
The scenario data storage unit 270 extracts scenario data for defining response information on a topic. For example, the scenario data storage unit 270 includes the ROM, the RAM, and the like of the Topic 20.

  As described later, in the present embodiment, scenario data consists of a plurality of statements. In Topiclet 20, by transitioning from one statement to another statement, it is possible to talk with the user while offering a topic to the user. The scenario data storage unit 270 stores a plurality of statements for advancing the conversation with the user. The specific example which advances a conversation with a user by making a statement transition is concretely demonstrated in FIGS. 5-14.

  As described later, the iWA 30 also includes a scenario data storage unit 320. The scenario data storage unit 320 of the iWA 30 stores all scenario data. On the other hand, the scenario data storage unit 270 of the Topic 20 may store it as a part of scenario data. It may be sent to Topict 20 as scenario data necessary for the user's conversation using Topict 20.

  In the present embodiment, when the user talks with the Topic 20, information such as an answer is not transmitted from the iWA 30 to the Topic 20 every time the user talks. When scenario data including a topic to be provided to the user is transmitted from the iWA 30, the scenario data is stored in the scenario data storage unit 270 of Topic 20. When the user has a conversation with the Topic 20, scenario data already stored in the scenario data storage unit 270 is used. In the present embodiment, since information such as an answer is not transmitted from the iWA 30 to the Topic 20 each time a conversation is made, the user can have a smooth conversation.

  Note that all scenario data stored in the scenario data storage unit 320 of the iWA 30 may be transmitted to the Topic 20. The topic can be provided to the user more smoothly.

<State control index storage unit 280>
The state control index storage unit 280 stores the state control index. The state control index is an index related to input information and response information. The state control index storage unit 280 includes the ROM, the RAM, and the like of the Topic 20.

  The Topic 20 includes a state control index storage unit 280, and the state control index storage unit 280 stores the state control index. In the present embodiment, the state control index is information that is not transmitted to the server (iWA 30 described later) but held in Topic 20. The Topic 20 determines response information with reference to the state control index stored in the state control index storage unit 280. By keeping the state control index in Topic 20, the amount of communication with iWA 30 can be reduced. In addition, since scenario data can be used with reference to the state control index in Topic 20, processing can be performed quickly, and conversation can be smoothly performed with the user.

<State control index>
In the present embodiment, three types of indicators, an input indicator, a progress indicator, and a character indicator, are used as state control indicators to control topics provided to the user. Other indicators may be used.

  The input index is information indicating what kind of input the user has made so far, that is, the history of the user's input. By using the input index, it is possible to guide the user to a situation that is easy to ask a question. That is, by using the input index, it is possible to provide the user with a topic based on the user's previous input. This enables the user to provide a topic to the user without asking unnecessary questions such as repeating similar questions.

  The progress indicator is information indicating what kind of topic has been provided to the user so far, that is, the history of the topic provided to the user. The progress indicator can maintain (store) a topic that the user wants to provide. This makes it possible to provide a topic to the user without causing the user to feel stress. By using the progress indicator, when providing a series of explanations as a topic, it is possible to resume the continuation of the series of explanations even if the user asks a question along the way.

  The character index is information indicating what kind of posture the user has input so far, that is, the history of the user's posture. For example, it is information indicating whether the user has actively entered or has been passively entered for a certain topic. In the positive case, it can be judged that the topic on a certain subject can be provided continuously. On the other hand, in the negative case, it can be determined that the user has to switch to a theme different from a certain theme to provide a topic.

  For example, it may be determined that users who may be interested in the car should continue to provide a topic related to the car. On the other hand, it can be determined that a user who is not interested in a car should be provided with a topic related to food and the like not related to the car.

  As described above, it is possible to switch the topic name (theme) of the topic provided to the user by the character index. The user can touch the topic belonging to the switched topic name without being aware of the topic name of the topic.

<IWA30>
In the present embodiment, the iWA 30 corresponds to, for example, hardware such as a server. The iWA 30 corresponds to the topic providing server 4 shown in FIG. The iWA 30 is communicably connected to the Topic 20. This is hardware for executing processing related to a topic provided to the user in Topic 20.

  Specifically, the iWA 30 includes a CPU (central processing unit), a ROM (read only memory), a RAM (random access memory), an HDD (hard disk drive), a display, a keyboard (all not shown), and the like.

  The iWA 30 includes an input information analysis unit 310 and a scenario data storage unit 320.

<Input Information Analysis Unit 310>
The input information analysis unit 310 analyzes the input information and generates input identification information. The input information is information input from the input unit 210 by the user. The input identification information includes a result of statistically analyzing the input information and information necessary for providing a topic from the result. For example, there is information such as the number and frequency of occurrence of the related words in the input information. Also, it includes scenario data (statement) and the like that are determined to be necessary for topic provision according to the result.

  Furthermore, analysis of the input information can analyze the intention and preference of the user from the question or the like input by the user. A relative analysis result can also be obtained by comparison with other user's input information and its input specific information. In addition, data such as an analysis dictionary can be generated in advance, and input information can be analyzed by the analysis dictionary.

  For example, the input identification information includes various pieces of information such as related words, scenario data, and the number of related words included in the scenario data. The scenario data includes topic information to be provided to the user and information on greetings necessary to have a conversation with the user.

<Related words>
Various data used in the topic providing system 1, the conversation control terminal 2 and the maintenance device 3 according to the present embodiment are configured on the basis of data referred to as related words. Unlike ordinary keywords used in ordinary search processing, the related words can associate various information such as history information and preferences with each other. The input information can be analyzed based on the related information held by the related words.

<Scenario data storage unit 320>
The scenario data storage unit 320 stores a plurality of scenario data. Here, the plurality of scenario data are all scenario data corresponding to a topic name required to have a conversation with the user in Topic 20. Among all the scenario data, scenario data determined to be necessary based on the input identification information is transmitted to the Topic 20. Therefore, when talking with the user in Topic 20, scenario data is not transmitted to Topic 20 each time the user talks. As described above, scenario data consists of multiple statements. Therefore, a plurality of statements constituting scenario data determined to be necessary based on the input identification information are transmitted to Topic 20.

  A statement is sent to Topic 20 as scenario data if it is determined that it is necessary based on the generated input identification information. If it is determined that it is not necessary, the statement already sent to Topic 20 is sufficient, and in this case, the statement is not sent to Topic 20.

  Incidentally, instead of transmitting only scenario data determined to be necessary based on the input specifying information to Topic 20, all scenario data stored in scenario data storage unit 320 may be transmitted to Topic 20. Good. Since all scenario data has already been transmitted to Topic 20, the time required to transmit and receive scenario data can be shortened, and conversation with the user can be smoothly advanced.

  Further, when switching the topic name, the scenario data stored in the scenario data storage unit 320 is rearranged, and scenario data corresponding to the topic name is transmitted to the Topic 20. That is, a statement corresponding to scenario data corresponding to the topic name is transmitted to Topic 20. The rearranged scenario data is stored in the scenario data storage unit 270 of the topiclet 20. This rearrangement of scenario data can be performed according to the topic name.

  It is also assumed that as the conversation with the user progresses, it becomes necessary to move from one topic name to another. In such a case, the scenario data corresponding to one topic name can not sufficiently cope. In such a case, it is possible to switch to scenario data corresponding to another topic name.

<IWA Manager40>
In the present embodiment, the iWA Manager 40 corresponds to, for example, hardware such as a server. The iWA Manager 40 corresponds to the maintenance device 3 shown in FIG. The iWA Manager 40 is communicably connected to the iWA 30. The iWA Manager 40 is mainly hardware for executing processing related to scenario data used in the iWA 30.

  Specifically, the iWA Manager 40 has a CPU (central processing unit), ROM (read only memory), RAM (random access memory), HDD (hard disk drive), display, keyboard (all not shown), and the like.

  The iWA Manager 40 has a scenario data editing unit 410, a scenario data verification unit 420, and a scenario data transmission unit 430.

<Scenario data editing unit 410>
The scenario data editing unit 410 is an apparatus or member that enables scenario data to be edited. The scenario data can be edited by the person in charge of the contractor of the topic providing system 1 operating a keyboard or the like. Editing includes adding, deleting, and changing scenario data. Specifically, editing is a process of adding, deleting, and changing statements that constitute scenario data.

  As new products are sold, new services are provided, various incidents occur, and users in new tiers increase, it is necessary to update scenario data to be able to respond to the latest topics. . For this reason, the person in charge can obtain various types of information via the network, and update the scenario data to the latest one based on the information. By updating the scenario data, it is possible to provide the user with a topic corresponding to the latest information.

  Also, the scenario data editing unit 410 can provide the user with a topic corresponding to the appropriate information by correcting the incorrect information such as a typographical error or a misprint or incorrect information.

<Scenario data verification unit 420>
The scenario data verification unit 420 is a device or a member that enables verification of the response of the scenario data edited based on the input identification information generated by the input information analysis unit. That is, the scenario data verification unit 420 is an apparatus or member for verifying whether the response of the scenario data edited by the scenario data editing unit 410 is appropriate.

  If the content of scenario data is appropriate, the response of scenario data will be appropriate. The scenario data in the present embodiment includes the output information output from the output unit 220, an output command for controlling the specification of the output to the output unit 220, a determination for controlling a statement, and a topic name. And control commands for changing state control indicators. Therefore, in addition to verifying whether the data output to the output unit 220 is appropriate, verification of whether the control of the output to the output unit 220 is appropriate, control of transition of scenario data, etc. It is necessary to verify whether it is appropriate.

  The scenario data verification unit 420 can generate input specification information using various input information assumed, and can verify the response of the scenario data edited by the scenario data editing unit 410. Therefore, since it is possible to verify whether or not the response of scenario data is appropriate for every user, it is possible to customize each of the users.

<Terminal device virtual construction unit>
The scenario data verification unit 420 can virtually construct the same environment as the Topic 20 by the terminal device virtual construction unit. By verifying scenario data in a virtual environment, it is possible to verify the output and operation of scenario data in an environment close to the environment actually used by the user, making it easy to determine whether the scenario data is appropriate or not And you can judge properly.

<Scenario data transmission unit 430>
The scenario data transmission unit 430 transmits the edited scenario data to the outside, for example, the iWA 30. In particular, scenario data edited by the scenario data editing unit 410 and further verified by the scenario data verification unit 420 is transmitted to the iWA 30. Therefore, the scenario data transmission unit 430 transmits the verified scenario data to the iWA 30.

  By doing this, scenario data transmitted from the iWA 30 to the Topic 20 can always be kept in an appropriate state. Therefore, a topic using appropriate scenario data can be provided to the user via the iWA 30.

<<< Data configuration provided by the topic provision server >>>
As shown in FIG. 15, the iWA 30 outputs scenario data, input specific information, and a topic introduction list. Hereinafter, an example of scenario data, an example of input identification information, and an example of a topic introduction list will be described.

<<< Example of scenario data >>
FIG. 14 shows an example of scenario data used in the topic providing system 1, the conversation control terminal device 2 and the maintenance device 3 according to the present embodiment. In the following, the scenario data shown in FIG. 14 will be described according to a specific processing procedure. The scenario data illustrated in FIG. 14 includes first to thirteenth plural statements. 5 to 12 are flowcharts showing the processing procedure of the first to thirteenth scenario data. FIG. 13 is a diagram illustrating an example in which the first to thirteenth scenario data are output to the output unit 220 by processing the scenario data.

  As described above, the scenario data (statement) used in the topic providing system 1, the conversation control terminal device 2 and the maintenance device 3 of the present embodiment includes output information, an output command, and a control command. The scenario data (statement) is composed of various elements such as output information, output command, and control command.

  FIG. 5 is a flow chart illustrating the processing of the first, second and third statements.

<First statement>
First, transition to the first statement. In FIG. 5, steps S511 to S521 correspond to the processing of the first statement.

  First, the progress index is set to -1 (step S511). This value indicates the progress status. Next, the output unit 220 is once erased (step S513), and the output unit 220 outputs “shift to a topic for eliminating anxiety” (step S515). By this processing, for example, as shown in FIG. 13A, a topic “worried about an earthquake” is input by the user's operation, and the text data display area of the output unit 220 It is displayed that “shift” (1311 in FIG. 13).

  Next, a predetermined image is output (step S517), and the process waits for 3 seconds (step S519). By this processing, as shown in FIG. 13A, the face image M1 of a predetermined color is displayed in the image data display area of the output unit 220, and waits for 3 seconds (1311 in FIG. 13).

  Next, the word "I'm worried." Is used as input information, and the topic name is switched to "topic material" (step S521). By this process, in this example, a transition is made to the second statement (1313 in FIG. 13).

<Second statement>
In FIG. 5, steps S523 to S525 correspond to the processing of the second statement.

  Transitioning to the second statement, first, monitoring of whether or not an absolute time, for example, 12:00, has been reached is started (step S523). Next, it is determined whether an absolute time has arrived (step S525) (1315 in FIG. 13). If the absolute time has not arrived (NO), transition is made to the third statement. If the absolute time has been reached (YES), the monitoring of the absolute time is canceled and a transition is made to the thirteenth statement (symbol EE).

<Third statement>
In FIG. 5, steps S527 to S533 correspond to the processing of the third statement.

  When transitioning to the third statement, it is determined whether the progress index is -1 (step S527) (1317 in FIG. 13). When it is determined that the progress index is -1 (YES), the process is transited to the fourth statement (symbol E1).

  If it is determined that the progress index is not -1 (NO), it is determined whether the progress index is -2 (step S529). If it is determined that the progress index is -2 (YES), the state transitions to the fifth statement (symbol E2).

  If it is determined that the progress index is not -2 (NO), it is determined whether the progress index is -3 (step S531). If it is determined that the progress index is -3 (YES), the state is transited to the sixth statement (symbol E3).

  If it is determined that the progress index is not -3 (NO), it is determined whether the progress index is -4 (step S533). If it is determined that the progress indicator is -4 (YES), the process is transited to another statement (sta: 200). If it is determined that the progress index is not -4 (NO), nothing is done.

  The second and third statements described above consist only of control commands. Thus, the statement may not have output information to be output to the output unit 220.

<Fourth and Seventh Statements>
FIG. 6 shows a process corresponding to the fourth and seventh statements. The fourth statement corresponds to steps S611 to S621. The seventh statement corresponds to steps S623 to S629.

<Fourth statement>
When transitioning to the fourth statement, measurement of relative time, for example, 120 seconds is started (step S611). Next, the input index is set to 1 (step S613), and the number of times the input index is set to 1 is counted (step S615).

  Next, it is determined whether the number of times the input index has been set to 1 has reached five (step S617) (1319 in FIG. 13). If the number of times of setting the input index to 1 has reached five (YES), transition is made to the tenth statement (reference E11).

  If the number of times of setting the input index to 1 has not reached 5 times (NO), it is judged whether the relative time, for example, 120 seconds has elapsed (step S619) (1319 in FIG. 13). If the relative time has elapsed (YES), the measurement of the relative time is ended (step S621), and a transition is made to a tenth statement (symbol E11).

  If the relative time has not elapsed (NO), transition is made to the seventh statement.

<Seventh statement>
When transitioning to the seventh statement, the output unit 220 is erased (step S623), and the output unit 220 has no question about "Worried about". Is output (step S625). Next, a predetermined image (for example, a face image M1 of a predetermined color) is output (step S627), and the process waits for 10 seconds (step S629). By this processing, as shown in FIG. 13B, the face image M1 of a predetermined color is displayed in the image data display area of the output unit 220, and waits for 10 seconds. Next, transition is made to the second statement (reference symbol ES).

<Fifth and Eighth Statements>
FIG. 7 shows the process corresponding to the fifth and eighth statements. The fifth statement corresponds to steps S711-S721. The eighth statement corresponds to steps S723 to S729.

<Fifth statement>
When transitioning to the fifth statement, measurement of relative time, for example, 120 seconds, is started (step S711). Next, the input index is set to 2 (step S713), and the number of times the input index is set to 2 is counted (step S715).

  Next, it is determined whether the number of times of setting the input index to 2 has reached five (step S717) (1319 in FIG. 13). If the number of times of setting the input index to 2 has reached five (YES), the processing transits to the eleventh statement (reference E12).

  If the number of times the input index is set to 2 has not reached five (NO), it is determined whether a relative time, for example, 120 seconds has elapsed (step S719) (1319 in FIG. 13). If the relative time has elapsed (YES), the measurement of the relative time is ended (step S721), and a transition is made to an eleventh statement (symbol E12).

  If the relative time has not elapsed (NO), transition is made to the eighth statement.

<Eighth statement>
When transitioning to the eighth statement, the output unit 220 is erased (step S723), and the output unit 220 has no question about "OK"? Is output (step S 725). Next, a predetermined image (for example, a face image M1 of a predetermined color) is output (step S727), and the process waits for 10 seconds (step S729). Next, transition is made to the second statement (reference symbol ES).

<Sixth and ninth statements>
FIG. 8 shows processing corresponding to the sixth and ninth statements. The sixth statement corresponds to steps S811 to S821. The ninth statement corresponds to steps S823 to S829.

<Sixth statement>
When transitioning to the sixth statement, measurement of relative time, for example, 120 seconds, is started (step S811). Next, the input index is set to 3 (step S813), and the number of times the input index is set to 3 is counted (step S815).

  Next, it is determined whether the number of times the input index is set to 3 has reached five (step S 817) (1319 in FIG. 13). If the number of times of setting the input index to 3 reaches five (YES), the process is transited to the eleventh statement (reference E13).

  If the number of times the input index is set to 3 has not reached five (NO), it is determined whether a relative time, for example, 120 seconds has elapsed (step S819) (1319 in FIG. 13). If the relative time has elapsed (YES), the measurement of the relative time is ended (step S 821), and transition is made to the twelfth statement (symbol E 13).

  If the relative time has not elapsed (NO), transition is made to the ninth statement.

<The 9th statement>
When transitioning to the ninth statement, the output unit 220 is erased (step S 823), and the output unit 220 has a question about “DANGER”? Is output (step S 825). Next, a predetermined image (for example, a face image M1 of a predetermined color) is output (step S827), and the process waits for 10 seconds (step S829). Next, transition is made to the second statement (reference symbol ES).

<Tenth statement>
FIG. 9 shows a process corresponding to the tenth statement. A transition to the tenth statement is made by the process (symbol E11) in FIG. 6 described above.

  When transitioning to the tenth statement, the progress indicator is set to -2 (step S911). Next, the output unit 220 is erased (step S 913), and the output unit 220 outputs “shift to next topic” (step S 915). A predetermined image (for example, face image M1 of a predetermined color) is output (step S917), and the process waits for 3 seconds (step S919). Next, the word "OK" is used as input information, and the topic name corresponding to this word is switched (step S921). Next, transition is made to the second statement (reference symbol ES).

  By the above-described process, as shown in FIG. 13D, the text data display area of the output unit 220 displays “shift to next topic”. By doing this, it is possible to transition from scenario data of one topic name to scenario data of another topic name.

<Eleventh statement>
FIG. 10 shows a process corresponding to the eleventh statement. A transition to the eleventh statement is made by the process of FIG. 7 described above (symbol E12).

  When transitioning to the eleventh statement, the progress indicator is set to -3 (step S1011). Next, the output unit 220 is erased (step S1013), and "output to next topic" is output to the output unit 220 (step S1015). A predetermined image (for example, face image M1 of a predetermined color) is output (step S1017), and the process waits for 3 seconds (step S1019). Next, the word "danger" is used as input information, and the topic name corresponding to this word is switched (step S1021). Next, transition is made to the second statement (reference symbol ES).

<Twelfth statement>
FIG. 11 shows a process corresponding to the twelfth statement. A transition to a twelfth statement is made by the processing in FIG. 8 described above (symbol E13).

  When transiting to the twelfth statement, the progress index is set to -4 (step S1111). Next, the output unit 220 is erased (step S1113), and "time is up" is output to the output unit 220 (step S1115). A predetermined image (for example, face image M1 of a predetermined color) is output (step S1117), and waits for 3 seconds (step S1119). Next, transition is made to a thirteenth statement (reference symbol EE).

<13th statement>
FIG. 12 shows a process corresponding to the thirteenth statement. A transition to the thirteenth statement is made by the processing of FIG. 11 described above (symbol EE).

  When transition is made to the thirteenth statement, the output unit 220 is deleted (step S1211), and “description is ended” is output to the output unit 220 (step S1213) (1321 in FIG. 8). A predetermined image (for example, face image M1 of a predetermined color) is output (step S1215), and the process waits for 3 seconds (step S1217). Then, transition to another statement (sta: 200).

  In the state shown in FIG. 13 (b), when the user inputs “After all, you are worried about politics” (FIG. 13 (c)), “politician” is displayed in the text data display area of the output unit 220. "" Does not lie, so there is no need to worry. "(Fig. 13 (c)) is displayed, and the state returns to the state shown in Fig. 13 (b) again.

<Structure of scenario data>
As in the first to thirteenth statements described above, in the present embodiment, scenario data consists of a plurality of statements. After transitioning to one statement and executing processing based on the one statement, transitioning to another statement is performed to execute processing based on the statement. By repeating such statement transition and processing in the statement, a topic can be provided to the user. As described above, scenario data (a plurality of statements) is used in the topic providing system 1, the conversation control terminal device 2 and the maintenance device 3 in the present embodiment. Here we describe the composition of a single statement.

  As in the first to thirteenth statements described above, the statement of the present embodiment is composed of various elements such as transition information and judgment. The statement can include elements necessary for providing a topic to the user and controlling a conversation with the user.

  Identification information is information attached to identify a statement. This identification information is referred to when transitioning a statement. One statement also includes transition destination information.

  That is, one statement includes both identification information and transition destination information. The identification information is identification information for identifying a statement, and is information indicating the statement itself. On the other hand, transition destination information is information for specifying a statement to be transited next. Identification information and transition destination information are information used when transitioning from another statement to one statement. That is, according to transition destination information defined in another statement, a statement of identification information that matches the transition destination information is searched, and transition is made to a statement (one statement) of identification information that matches the transition destination information. In this way, statements can be made to transition one after another by using both identification information and transition destination information.

  Judgment consists of judgment based on index and judgment based on time. The judgment based on the index is a judgment to determine whether the index satisfies a predetermined condition. When the index satisfies a predetermined condition, it is determined to be true, and when the index is not satisfied, it may be determined to be false and branch. The determination based on time is a determination to determine whether the time satisfies the predetermined condition. When the time or time satisfies the predetermined condition, it is determined as true, and when the predetermined condition is not satisfied, it is determined as false and it is possible to branch.

  The output information is text data to be output to the output unit 220. An image can also be output to the output unit 220 according to the output information by including identification information (for example, a file name or the like) for specifying the image data of the output information.

  The setting element is an element for setting a state index. Statements can be transitioned or branched based on the set state indicator.

  Other elements include, for example, line feed and deletion. The line feed is for outputting a line feed of the text in the output unit 220. The erasure is for erasing the text or image output to the output unit 220.

  The output control element is an element for controlling the output of the output information described above. For example, it is possible to specify a time for outputting output information, or to specify an image to be output together with the output of output information.

  The transition destination information is information for specifying the statement of the next transition destination. The identification information that matches this transition destination information is searched, and transition is made to the statement of the identification information.

  The topic switching information is an element for switching the topic name. Scenario data (a plurality of statements) of the present embodiment is defined for each of a plurality of topic names. That is, scenario data is defined such that scenario data corresponds to each of a plurality of topic names. That is, in the present embodiment, scenario data is defined to correspond to each of a plurality of topic names, and each of the scenario data is composed of a plurality of statements.

  If one topic name is sufficient for conversation with the user, the statements may be transitioned one after another using scenario data corresponding to one topic name. However, it is also assumed that it becomes necessary to move from one topic name to another as the user's conversation progresses. In such a case, the scenario data corresponding to one topic name can not sufficiently cope. For this reason, by defining scenario data corresponding to each of a plurality of topic names in advance, even if it is necessary to move the topic from one topic name to another topic name, the other topic names are supported. It can be switched to scenario data. Scenario data corresponding to other topic names is also composed of a plurality of statements. By sequentially transitioning statements of scenario data corresponding to other topic names, it is possible to advance a conversation with the user for other topic names.

  When switching the topic name, all scenario data stored in the iWA 30 is rearranged to generate scenario data corresponding to the topic name. The rearranged scenario data is stored in the scenario data storage unit 270 of the topiclet 20. This rearrangement of scenario data can be performed by determining a combination of a plurality of statements according to the topic name.

  In addition, it is preferable to determine whether to switch the topic name based on the above-mentioned character index. The character index is information indicating whether the user is positive or negative about a certain topic. In the positive case, it can be determined that the topic can be continuously provided without changing the topic name. On the other hand, in the negative case, it can be determined that the topic name should be switched to provide a topic.

<< Example of input specific information >>
Identification information corresponding to the input information or information for specifying the identification information is added to the input identification information, and scenario data can be activated by this information.

  In FIG. 15, the input specifying information in FIGS. 1 to 4 also includes a topic introduction list to be described next.

<< Example of the topic introduction list >>
FIG. 16 is a diagram showing an example of the configuration of the topic introduction list.

<Composition of topic introduction list>
The topic introduction list is a list of related word sets and topics. The related word set is a set of related words included in the topic. The neighborhood of a related word is a set of related words including the related word A when focusing on a certain related word A. The neighborhood system of related words is a set of neighborhoods of related words. The number of neighborhoods of related words is the number of elements in the neighborhood system of related words. The phase of related words can be found by looking at the neighborhood system of related words. By performing topic analysis, it is possible to display neighborhood systems for all sets of related words. By performing preference analysis, it is possible to display the neighborhood system of related words in order of preference. The scenario related to the topic introduction is a scenario that can be configured based on the proximity and connection of topics based on the neighborhood system of related words.

  The topic is the text to which the action and the index are assigned. An action is a change activated by clicking a text or the like. The topic has a set of related words as an index. When you click the index, related words of related words included in the index are displayed.

<Use of topic introduction list>
FIG. 17 shows that the topics of the topic introduction list are connected by the related words into which related word structures such as a related word dictionary and a preference dictionary are introduced. The user can detect topics by focusing on how the topics in the topic introduction list are connected.

  Also, the user can detect a topic from various points of view by performing topic switching based on the related words while referring to the related word structure in the topic introduction list.

Topic switching by related words
In scenario data that defines response information related to a topic, it is necessary to switch and introduce the topic itself as well as the topic switching by topic name. The topic analysis provides "relevant verbs and relevant verb structures" necessary to switch the topic itself in the scenario data through the topic introduction list.

  As shown in FIG. 17, the closeness and connection of the topic can be seen by the relative word, so that the topic itself can be switched by using the relative word in the scenario data. For example, with respect to the current topic, topic switching to a topic close to the content or topic switching to a topic in which the content is connected can be performed.

  In addition, as the related words structure is introduced into the related words, use variable related words (for example, the most frequent related words, the most popular related words, etc.) in which the related words are variableized in the scenario data Enables various introductions of topics.

  By being able to switch the topic by using the related words, the topic providing system 1 can reproduce the manual work in which the user detects the topic using the topic introduction list as the topic providing service using the scenario data.

<<< Topic analysis >>>
Next, the topic analysis unit provided in the maintenance device 3 in FIG. 4 will be described again based on FIGS. 15 to 18. As described above, the maintenance device 3 constitutes a scenario data verification unit for creating in advance the information that the contractor of the topic providing system 1 wants to provide to the user. And although there is no need to create information (topic introduction list) to be provided to the user, the maintenance device 3 of this embodiment can be implemented by adding the state control index storage unit described above to the scenario data verification unit. It is configured to also function as a terminal device virtual construction unit for virtually functioning as the conversation control terminal device 2. That is, a terminal device virtual construction unit is configured by adding the state control index storage unit to the scenario data verification unit, and the terminal device virtual construction unit corresponds to the topiclet 20 and the iWA Manager 40 described above.

  As illustrated in FIG. 15, according to the topic analysis unit, it is possible to analyze a topic and to perform an output for visualizing a topic introduction list. That is, the topic analysis unit is configured to execute generation of a topic list to which the proximity and connection of topics are given via the related words that relate topics. In addition to the topic analysis unit, maintenance device 3 also uses a topic list and the related words to introduce a topic introduction list to the user (corresponding to the topic list in FIG. 17), and the user's It is also characterized by having a scenario data editing unit for making an input related scenario for responding to an input editable as the scenario data. In the case of the maintenance device 3, since the conversation control terminal device 2 is virtually constructed, the user's input here corresponds to the input of the simulator (person in charge).

  The following describes topic analysis in the topic analysis unit described above, generation of a topic introduction list, and scenario data editing by the scenario data editing unit.

<<< output of response based on topic analysis >>>
FIG. 15 is a diagram showing a process of generating response information based on topic analysis and outputting the response information to an output unit. FIG. 16 is a diagram showing an example of the configuration of the topic introduction list. FIG. 17 is a diagram showing a process of topic extraction, generation of related word dictionary, and generation of preference dictionary. The topic analysis is executed by the topic analysis unit to build the iWA 30 preference dictionary and the related word dictionary as a server. The control unit configured by the CPU or the like of the maintenance device 3 uses these dictionaries to add a plurality of related words included in the topic to each of a plurality of topics (topic material) input by the contractor's person in charge. Apply automatically.

  FIG. 15 is a diagram showing a process of generating response information based on topic analysis and outputting the response information to an output unit.

  As a result of topic analysis, the iWA 30 generates scenario data from input related scenarios and topic introduction scenarios. The iWA 30 generates input specific information from elements such as identification information. The iWA 30 generates a topic list from a set of related words and topics.

  Then, response information is generated based on the generated scenario data, the input identification information, and the topic introduction list, and the output unit 220 outputs the response information.

  By doing this, response information can be generated from topic analysis and response information can be output. By generating response information by topic analysis, the response information used in the conversation with the user can be made appropriate for each of the users, and the conversation with the user can be made smoother.

<Composition of topic introduction list>
FIG. 16 is a diagram showing an example of the configuration of the topic introduction list.

  The topic introduction list is a list of related word sets and topics. The related word set is a set of related words included in the topic. The neighborhood of a related word is a set of related words including the related word A when focusing on a certain related word A. The neighborhood system of related words is a set of neighborhoods of related words. The number of neighborhoods of related words is the number of elements in the neighborhood system of related words. The phase of related words can be found by looking at the neighborhood system of related words. By performing topic analysis, it is possible to display neighborhood systems for all sets of related words. By performing preference analysis, it is possible to display the neighborhood system of related words in order of preference. The scenario related to the topic introduction is a scenario that can be configured based on the proximity and connection of topics based on the neighborhood system of related words.

  The topic is the text to which the action and the index are assigned. An action is a change activated by clicking a text or the like. The topic has a set of related words as an index. When you click the index, related words of related words included in the index are displayed.

<Extraction of topic, generation of related words dictionary, generation of preference dictionary>
FIG. 17 is a diagram showing creation of a topic introduction list and profiling of a user using a preference dictionary.

  The topic introduction list is visualized in a display form as shown on the screens (symbols 1813 and 1811) of the display device provided in the maintenance device 3 in FIG. 17, and is obtained from the related vocabulary dictionary F or an external news source. The topic list G is extracted from the iWA 30, and a topic is automatically added by the control of the maintenance device 3 based on the topic list G consisting of a manual input of the topic by the person in charge and a topic data group introduced from the outside It will be built.

  First, the person in charge of the contractor is the input keyboard as the input device of the maintenance device 3 as the input source of the content to be input and displayed as the topic (field of topic information setting on screens 1813 and 1811) in the topic introduction list The first form in which the topic material is directly input and set, and log data collected from the outside by the iWA 30, for example, data that can be collected via networks such as Twitter and blog, the person in charge of the input device of the maintenance device 3 There is a second form in which the iWA 30 automatically extracts the topic material based on the topic input by the user (1811). From the extracted topics, data of a topic browsing list to be a candidate for a topic introduction list is generated. The iWA 30 refers to the related-word dictionary F to display a topic manually input by the person in charge via the maintenance device 3 on the screen 1815, to the topic in which a plurality of related words are input under the control of the iWA 30. Next, with reference to the related step dictionary associated with the topic manually input by the person in charge as a key, the iWA 30 refers to the topic list G consisting of general news groups collected from the outside, and the person in charge The iWA 30 automatically extracts an external topic related to the input topic, and the maintenance device 3 receives the extracted data and displays the data as shown on the screen 1811.

  Then, the person in charge automatically performs topic reduction and addition while automatically inputting the topic by means of a direct input device such as an input keyboard and the topic list database in which the iWA 30 has accumulated information from the outside beforehand. Build an introductory list.

  On the other hand, in addition to the generation of the topic introduction list described above, the maintenance device 3 of the present embodiment refers to the preference dictionary E obtained from the iWA 30, and based on the related-word dictionary F for the input topic. The related words of data of the user type list in which related words are displayed on the linked screen 1815, the user ID generated from the response history of another user, the user type, and the related words are compared. By this comparison, for example, as shown on the screen 1817, “the most scary” extracted as a related word to a certain topic refers to the preference dictionary E constructed with the other user history in the past, It is possible to analyze and display whether information related to the same type of user is associated with the information input by the user of the user type for visualization.

  This analysis result is stored together with the user ID for specifying the user who has input the same related words, the user type (for example, the customer of yesterday), and the related words common to all the users, and based on the stored data Can be used to analyze user preferences. As the method of use, for example, for a user who is considered to have the same preference, a specific service that is considered to match the preference can be specified from the user ID, and a delivery destination such as the user's email address can be specified. It becomes possible to specify the delivery destination and distribute it, or to provide a topic other than the service that suits the preferences to the identified delivery destination.

  As shown in FIG. 18, the topic is input to the topic topic data setting screen by the manual input according to the first form of the maintenance device 3 described above, and the topic introduction list is generated by the related word dictionary and the topic list to make iWA 30 as a server. Shows the flow of processing up to output. Although FIG. 17 illustrates the process of performing the profile of the user from the topic input using the preference dictionary, the process of performing this profile is different from the process of generating the topic introduction list, so in the flow of FIG. The description of the process of performing the profile of the user will be omitted from the above-mentioned topic.

  Referring to FIG. 18, the control unit of maintenance device 3 displays a setting screen of the topic material on the display screen of maintenance device 3 and waits for an input of a topic to be executed by the person in charge (S2000).

  Next, when it is determined that a topic has been input via an input device such as a keyboard on the input screen and a switch for setting topic information has been input (S2001: YES), the topic data is sent to the iWA 30 as a server. The set topic material is transmitted, and in accordance with the content of the transmitted data of the topic material, one or more related words are extracted from the related word dictionary F under control of the CPU of the iWA 30. In the case of the screen 1813 of FIG. 17, “panic discussion topic” is transmitted from the maintenance device 3 to the iWA 30 as a topic topic, and based on the content of the “panic discussion topic”, it is called “panic” based on the control of the iWA 30 CPU. Related words are extracted. Then, when the related topic is extracted in advance from the data of the topic list G existing in the database of the iWA 30 by using the related word "panic" as a key, as illustrated in the screen 1813, the topic list (the number of selected topics : 1424 total number of topics 1424) Thru Butt: 17.25 is displayed. What can be found from this screen 1813 is that 1424 topics are extracted under the control of the CPU of the iWA 30 by using the related word "panic" as a key, and 1424 items which may eventually constitute a topic introduction list The topic and the plurality of related words associated with the topic are received by the maintenance device 3 and illustrated on the screen 1813. To explain briefly, list data in which a plurality of topics and a plurality of representative associated words are associated with each topic using the associated word "panic" as a key for displaying the screen 1813 of FIG. 17 Is received from the iWA 30 (S2002).

  Such an input was generated and displayed as 1424 topics available as topic introduction list candidates with the first related word "panic" as a key, but if this number is large, it is visualized The person in charge can determine the content of the topic, and narrowing-down can be performed so as to be a topic to be adopted for the topic introduction list that is more targeted. That is, referring to the flow of FIG. 18, timeout is made after a predetermined time elapses and the input of the additional topic material in step S 2001 is executed again without confirming the list data of the topic introduction list in step S 2004. Can. For example, on the screen 1815, the input representative related words and the received data to be the topic are displayed by inputting the second topic topic “I'm relieved”, and there is a selection Topic number 136 Total topic number 1424) Throughput: It is displayed as 16.54. That is, the list data being generated as the topic provision list is reduced from 1424 to 136. By this, it is changed to the form close | similar to the topic provision list | wrist which followed the said person in charge.

  Furthermore, when the person in charge can not think of a new topic, the display form can be changed to a display form centered on the related words. In this display form, as shown in the screen 1815, by inputting “condition setting: priority related words” by the operation of the input device, it is possible to display related words in a list form giving priority. . That is, as shown in the screen 1815 of FIG. 17, a plurality of related words given under the control of the CPU of the iWA 30 with respect to the second topic are 136 topics (1, 2, 3 · · ·). · · · It is possible to switch to a display form in which a plurality of related words are displayed in a plurality of columns in each row 136). In the switching of the display mode, as described above, the control unit (not shown) of the maintenance device 3 enumerates the related words for each topic displayed for each row in the column direction according to the screen switching input of the input device. Change the display to the form. The person in charge can determine, from the plurality of related words displayed on the screen 1815, related words suitable for topic introduction to be created by the person in charge from the display mode, and can input the related words with the input device (S2001: YES). The screen 1811 shows this input state, and in this case, the related word "most scary" is input as a new topic material. Then, referring to the database of topic lists under the control of the CPU of the iWA 30, the related words “the most scary” are referred to, and based on the input topic material, a new topic introduction list consisting of a plurality of related words and topics. The list data as a candidate is received (S2002) and displayed (S2003).

  The topic list G is a list including information collected by an external information collection unit such as the Internet from the outside by the iWA 30 as a server. Then, in each topic of the topic list G, a plurality of related words are associated in advance and stored in the database of the iWA 30. Thus, in this embodiment, there is a possibility that the variation for setting the topic may become obsolete due to lack of knowledge or ability of the person in charge, and other related topics obtained from the iWA 30 topic list as a server To be visualized by displaying the topic itself and a plurality of related words related to the topic on the screen of the maintenance device 3, as described above based on the topic obtained from these topic lists The possibility of being able to make the variation of the topic transitioning to

  As for the modification of the topic provision list candidate as described above, the input completion for input completion is made (S2004: YES), and when a predetermined time elapses, a timeout occurs, and an input screen for inputting the above-mentioned topic is displayed. After waiting for the next topic, the processing of S200 to S2003 is sequentially repeated.

  Then, when the input completion is input (S2004: YES), data as a topic introduction list is output to the iWA 30 as a server (S2005).

<<< Overview of technical ideas in information retrieval system >>>
The information search system of the present invention provides the user with a keyword (character string) capable of acquiring the latest unknown topic by using the mechanism of the topic providing system 1 as shown in FIG. First, an overview of the technical concept in the information search system of the present invention will be described with reference to FIG.

  In the information search system according to the present invention, an important character string (specification that can identify the topic through the sentence analysis process 511, the preference analysis process 512, and the topic analysis process 513 from the external log 502 including personal information generated by the individual) The relationship (character string) and the distribution status can be displayed, and the user can understand the topic by viewing the display. From the input of the external log 502 to the provision of the display of a specific string is performed instantaneously. As an example of a specific string, a relative is used here.

  For example, when the result of searching a web page or the like on the Internet is regarded as topic information, a topic dictionary which is a summary of topic information can be obtained by compressing and summarizing the topic information. In the information search system of the present invention, the compression of the topic dictionary is performed before the above-described sentence analysis processing 511. For example, only text data that can be topic information, excluding tags and script sentences, is extracted from search results of web pages etc., and the text data extracted as such is to be processed by the sentence analysis process 511. .

  In a more specific example, company information (topic information) is information related to a company, which is text data generated by an individual, and this is a knowledge space related to a company. This knowledge space is compressed, summarized, and converted into a partial knowledge space without using a language dictionary in the above-described processing. As a result of this processing, a dictionary (a partial knowledge space related to a company) consisting of a set of decomposed text data (related words) that can identify company information is obtained. This partial knowledge space contains information representing the connection between related words, as described later.

  In the conversation control terminal 2 ′ ′ (for example, from the user of the conversation control terminal 2 ′ ′), a condition for collecting the external log 502 is given, and the above-described processing (state analysis processing 511, preference analysis processing 512, As a processing result of the topic analysis processing 513), the related words are provided to the conversation control terminal device 2 ''. The conversation control terminal device 2 ′ ′ is, for example, a device such as a PC (personal computer), a smart phone, or a robot, and if the conversation control terminal device 2 ′ ′ is a PC, the related term of the processing result is the conversation control terminal device 2 They are displayed on the '' display, and they are provided to the user of the conversation control terminal device 2 '' as information for instantly grasping the topic. The conversation control terminal device 2 ′ ′ is configured as a modification of the above-mentioned conversation control terminal device 2 and the conversation control terminal device 2 ′.

  The sentence analysis process 511, the preference analysis process 512, and the topic analysis process 513 described above are performed by the topic providing server 4 '. The topic providing server 4 ′ is configured as a variation of the above-described topic providing server 4.

<< Description of outline of sentence analysis process >>
The sentence analysis process 511 analyzes sentence information included in the external log 502 based on the appearance characteristic of the character string, and selects a related word 503.

  The sentence analysis process 511 selects (extracts) related terms that can identify the topic from the external log 502 without using previously stored and prepared dictionary data such as morpheme data. That is, the common character strings appearing in the external log 502 are searched, and the related words are extracted according to the degree of difference between adjacent characters immediately before those character strings and the degree of difference between adjacent characters immediately after them.

  The external log 502 may be an episodic information (for example, data stored in a predetermined log format, web pages (home pages) and blog text data published on the Internet), and TWITTER (registration). (Trademark information on trademarks), data generated and edited in advance by an arbitrary organization, and text information in a database. In addition, various data may be used, such as text data acquired from the audio file / moving image file through voice recognition processing.

  Also, the external log 502 is data collected by the collection condition. For example, it may be text data described in a web page (home page) 501 shown as a search result of a keyword search, a sentence described in the blog of a user having a certain attribute, tweet information in TWITTER, etc. . The search conditions and the like in the keyword search can be designated by the user from the conversation control terminal device 2 ′ ′. Note that one external log 502 may be one including a plurality of text files (for example, one including a plurality of WEB pages (HTML files) included in one web site) or one text file It may be a divided part (for example, one of the texts contained in one file divided into 10,000 lines).

<< Description of outline of preference analysis process >>
The preference analysis process 512 catches how to use the related words extracted by the sentence analysis process 511 based on the internal log 506, and determines their importance.

  The internal log 506 is data indicating the preference of the user (including the organization or organization to which the user belongs), and is, for example, data stored in a predetermined log format. The internal log 506 includes, for example, data indicating what kind of related words are likely to be used by the user. Thus, in the present specification, a related term associated by the importance according to the user's preference is referred to as a topic key (cluster) 504.

<< Description of outline of topic analysis processing >>
The topic analysis processing 513 captures the distribution of the topic key 504 generated by the preference analysis processing 512 based on the topic material 507, and provides the user with the distribution of related words associated with each other.

  As described above, the topic material 507 is set when the contract person's person in charge directly inputs and sets the input device of the maintenance device 3, and the person providing the topic provision server 4 ′ is the input device of the maintenance device 3. There is a case where it is automatically extracted from an external log 502 collected from the outside (for example, data that can be collected via a network such as a TWITTER or a blog) based on the input keyword.

  By such topic analysis processing 513, it is possible to indicate how the related words are distributed in the topic, and it is possible to recommend the related words according to the user of the conversation control terminal device 2 ''.

<<< Overview of Information Retrieval System >>>
Next, an overview of the information search system will be described with reference to FIG. An information search system 100 shown in FIG. 20 includes a conversation control terminal device 2 ′ ′ and a topic providing server 4 ′, and a predetermined network (LAN, etc.) is provided between the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′. It is connected by the Internet, WAN, wireless communication etc.).

<< Overview of Conversation Control Terminal Device 2 ''>>
The conversation control terminal 2 ′ ′ includes an input control unit 21, a search control unit 22, a transmission control unit 23, a reception control unit 24, a response information determination unit 25, an output control unit 26, and a network interface (I / F) unit 27. including. The scenario data 28 is stored in a main storage device such as a RAM or an external storage device such as a hard disk or a semiconductor memory.

  The input control unit 21 receives an input from the user of the conversation control terminal device 2 '' using a keyboard, a mouse or the like, and delivers input data or the like to the corresponding function unit according to the content of the input. For example, the user inputs a search keyword by using a keyboard, or clicks on a display area of related words of interest by using a mouse.

  The search control unit 22 includes a general web browser operating on the conversation control terminal device 2 ′ ′. The conversation control terminal device 2 ′ ′ is connected to, for example, the Internet, and when the user operates the WEB browser to search a WEB page (generally available Internet search), the search control unit 22 The obtained search result is transmitted to the transmission control unit 23. The search results include the address (for example, Internet address identification information such as a URL) of the web page related to the search keyword.

  When the transmission control unit 23 receives the search result from the search control unit 22, the transmission control unit 23 transmits this as input information to the input information analysis unit 41 of the topic providing server 4 ′ by, for example, API transmission.

  The reception control unit 24 receives input identification information and the like transmitted from the input information analysis unit 41 of the topic providing server 4 ′, and supplies this to the response information determination unit 25.

  The response information determination unit 25 determines response information based on the scenario data 28 and the input identification information. The input information analysis unit 41 determines response information based on input specific information (for example, data for displaying the distribution of related words) and scenario data determined to be necessary for the display.

  The output control unit 26 controls the conversation control terminal device 2 ′ ′ to display the response information determined by the response information determination unit 25.

  The network interface unit 27 accesses and exchanges data with the topic providing server 4 ′ connected via the network, and other computers (eg, a server having an Internet search engine connected via the Internet, etc.) Control access and data transmission / reception between

  The scenario data 28 is data for defining response information on a topic provided to the user, as shown in FIG. The scenario data 28 is data stored in advance in scenario data 55 of a topic providing server 4 'described later. The scenario data judged to be necessary based on the input specific information generated by the input information analysis unit 41 of the topic providing server 4 ′ is extracted from the scenario data 55, and the extracted scenario data is a conversation control terminal device It is stored in 2 ′ ′ scenario data 28. The scenario data extracted from the scenario data 55 is stored in the scenario data 28 by the processing of the reception control unit 24 and the response information determination unit 25.

  The scenario data 55 of the topic providing server 4 'stores all scenario data, and based on the information defining the transition to different topics, scenario data reassembled from the scenario data of the scenario data 55 is generated and reorganized Only the corrected scenario data can be stored in the scenario data 28 of the conversation control terminal 2 ''.

<< Overview of Topic Providing Server 4 '>>
The topic providing server 4 ′ includes an input information analysis unit 41 and a network interface (I / F) unit 47. In addition, search result data 48, related word candidate data 49, related word dictionary 50, preference data 51, related word / co-occurrence word data in a main memory such as a RAM, or an external memory such as a hard disk or a semiconductor memory. 52, topic data 53, comparison result data 54, and scenario data 55 are stored. Note that these data can take various data formats and data storage formats.

  The input information analysis unit 41 analyzes the input information received from the conversation control terminal device 2 ′ ′ and generates input specific information. The input identification information is information generated as a result of analyzing various types of information contained in the input information, and includes, for example, the distribution of related words described later. The input information analysis unit 41 further includes an external log acquisition control unit 42, a sentence analysis unit 43, a preference analysis unit 44, a topic analysis unit 45, and an information update unit 46.

  The external log acquisition control unit 42 indicates that the input information received from the conversation control terminal device 2 '' is identification information for identifying the external log 502 (for example, a search result including the address of the web page 501 related to the search keyword) In this case, for example, the identification information is accessed via the Internet, and the corresponding HTML data and the like are acquired. Further, when the input information received from the conversation control terminal device 2 ′ ′ includes the text data itself for which the related words are to be extracted, the data is provided as the external log 502 to the sentence analysis unit 43.

  When information identifying the external log 502 (for example, a search result including the address of the web page 501 related to the search keyword) is received from the crawler (see FIG. 39) automatically started at a predetermined interval, the information The external log 502 is accessed, the corresponding data is acquired, and the acquired data is provided to the information updating unit 46 for comparing related word dictionaries.

  The sentence analysis unit 43 acquires text data from the external log 502 acquired by the external log acquisition control unit 42, extracts important relevant words included in the text data according to the appearance characteristic of the character string, It is stored in the dictionary 50.

  The preference analysis unit 44 determines the importance of the related words stored in the related word dictionary 50 by the sentence analysis unit 43 based on the preference data 51, and stores the determination result in the related word / co-occurrence word data 52. The preference data 51 is data including an internal log 506 storing the usage of the related words by the user.

  The topic analysis unit 45 captures the distribution of the related words stored in the related word / co-occurrence word data 52 generated by the preference analysis unit 44 based on the topic data 53, associates the related words with each other, and the related words Update the co-occurring word data 52. The topic data 53 is data including a topic material 507 input or set automatically by the contractor's person in charge or extracted automatically.

  The information updating unit 46 selects related words based on text data collected under different collecting conditions, generates a related word dictionary, compares the related word dictionaries generated in this way, and compares comparison results with comparison data Store in 54.

  The network interface unit 47 accesses and receives data from the conversation control terminal device 2 ′ ′ connected via the network, and other computers (eg, a server including an Internet search engine connected via the Internet, etc.) Control access to and between

  In this embodiment, the information search system 100 is described as a system including the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′. However, the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′ It can also be configured as a single integrated computer. Conversely, similar functions can be distributed to three or more networked computers. Further, the sentence analysis unit 43 can be configured as one independent sentence analysis device, and similarly, the information updating unit 46 can be configured as one independent information update device. The sentence analysis device and the information update device can be connected to other devices via a network to configure the information search system 100 described above.

<Outline of sentence analysis unit>
Next, the outline of the sentence analysis unit 43 will be described with reference to FIG. In the sentence analysis unit 43, the same character string is searched from the text data, and the degree of difference between the previous adjacent character and the degree of difference between the subsequent adjacent characters are determined for each of the searched same character strings, and the determined differences Based on the degree, it is determined whether or not the retrieved “same string” is a relevant word that is highly relevant with respect to the topic and that text data can be semantically identified. The degree of difference between the preceding adjacent characters is an index indicating how different the characters appearing immediately before the retrieved “same character string” are. Similarly, the degree of difference of the subsequent adjacent characters is an index indicating how different the characters appearing immediately after the retrieved “same character string” are different.

  Here, for example, among the “same character string” searched, a character string in which the degree of difference between the preceding adjacent character and the degree of difference between the subsequent adjacent character is large is determined as the associated word. The one or more character strings thus determined are stored in a predetermined storage means, as needed.

  In such character string extraction, when attention is focused on a plurality of identical character strings included in text data, characters of many variations appear as characters located immediately before each character string, and each character string In the case where the appearance characteristic that characters of many variations appear as characters immediately after is recognized, it is based on the idea that the character string is an independent and commonly used term. Thus, in the sentence analysis unit 43 of the present invention, a character string is extracted based on the idea based on the kinematics of the character string.

  Here, assuming Japanese text data in which the character string "Iroha" appears 100 times, 100 characters of this character string "Iroha" are searched, and for each character string "Iroha", the character immediately before is what Examine. As a result, it is assumed that the fact that 30 characters including "A" and "I" appear will be obtained. This indicates that, for example, expressions such as "... Airoha ..." or "... ... Yoroha ..." exist in the above-mentioned Japanese text data. On the other hand, for each string "Iroha", the character immediately after is examined. As a result, it is assumed that the fact that 20 characters including "wa" and "n" appear will be obtained. This indicates that, for example, expressions such as "... Irohawa ..." or "... Irohan ... ..." exist in the above-mentioned Japanese text data.

  In this case, the degree of difference between the preceding adjacent characters is determined based on, for example, the number of 30 variations including "A" and "I". The degree of difference between the subsequent adjacent characters is, for example, It is determined based on the number of variations of 20 ways including "N". Here, when it is determined that the degree of difference between the previous adjacent character and the degree of difference between the subsequent adjacent character are large, the characters before and after the character string "Iroha" are changed with great diversity, thereby It is determined that the column "Iroha" is likely to be an independent term and a high importance term, determined as a related term, and stored in storage means as needed. Whether the degree of difference between the preceding adjacent character and the degree of difference between the subsequent adjacent character is large is determined by common or individual judgment criteria.

  The sentence analysis unit 43 includes a text data acquisition processing unit 43a, a character string search processing unit 43b, a difference degree determination processing unit 43c, and a related term determination processing unit 43d. Further, the related word determination processing unit 43d includes a related word determination unit 43d-1 and a ranking management unit 43d-2.

  The text data acquisition processor 43a acquires the external log 502 (text data to be processed) and provides this to the character string search processor 43b (text data acquisition processing 520 shown in FIG. 25 described later). The character string search processing unit 43b performs a character string search process 530 shown in FIG. The difference degree determination processing unit 43c performs the difference degree determination process 540 shown in FIG.

  The related term determination processing unit 43d determines related terms, and stores the determined related terms in the related term dictionary 50 as necessary (related term determination processing 550 shown in FIG. 25).

  Further, the related term determination unit 43d-1 determines whether or not the same character is a related term from the degree of difference between adjacent characters before and after the same character string included in the external log 502. When a plurality of related words are determined in one external log 502, the ranking management unit 43d-2 ranks the related words as needed.

<Overview of Information Update Department>
Next, the outline of the information updating unit 46 will be described with reference to FIG. The information updating unit 46 includes a text data acquisition processing unit 46a, a character string extraction processing unit 46b, a dictionary comparison processing unit 46c, and a comparison result output unit 46d.

  The text data acquisition processing unit 46a acquires the external log 502 (text data to be processed) and provides this to the character string extraction processing unit 46b (text data acquisition processing 700 shown in FIG. 39 described later). The character string extraction processing unit 46b extracts related words from the external log 502, and stores them in the corresponding related word dictionary 50 (character string extraction processing 710 shown in FIG. 39). The character string extraction processing unit 46 b is, for example, the same processing as the processing by the sentence analysis unit 43 described above.

  The dictionary comparison processing unit 46c compares a plurality of related word dictionaries 50, and stores the comparison result in the comparison result data 54 (dictionary comparison process 720 shown in FIG. 39).

  The comparison result output unit 46e acquires the comparison result to be displayed from the comparison result data 54, and transmits the input identification information including the same to the conversation control terminal device 2 ''.

<<< Description of an embodiment in which the information search system of the present invention is applied to FAQ search >>>
Next, using the information search system according to the embodiment of the present invention, a FAQ search system will be described which displays the result of the FAQ search according to the user's instruction.

<< Screen transition of FAQ search system >>
FIG. 23 shows the screen transition of the FAQ search system. First, the user gives a predetermined instruction on the conversation control terminal 2 ′ ′, causes the display to display the FAQ search screen 600, and inputs a desired search keyword (using a keyboard or the like) there. The FAQ search screen 600 is, for example, an input instruction screen as shown in FIG. 35A. On the FAQ search screen 600, a search keyword input unit 601 and a “FAQ search” button 602 are displayed.

  Here, when the user inputs a search keyword ("Network" in the example of FIG. 35A) in the search keyword input unit 601 and clicks the "FAQ search" button 602 with a mouse etc., the FAQ candidate display screen 610 Is displayed. The FAQ candidate display screen 610 is, for example, a display screen as shown in FIG. 35 (B), and the related vocabulary index display portion 611, the candidate question sentence display portion 612, and the “Return to FAQ search screen” button 613 are displayed. ing. The questions shown in the candidate question sentence display portion 612 all relate to "network", and the result of the search based on the search keyword input by the user is displayed. The set of related words indicated in the related word index display portion 611 is a set of related words included in the corresponding question. Here, when the user clicks the "return to FAQ search screen" button 613, the display on the display of the conversation control terminal 2 "returns to the FAQ search screen 600.

  In the FAQ candidate display screen 610, when the user selects one of the candidate question sentences displayed in the candidate question sentence display unit 612 by clicking the mouse (arrow (1)), the FAQ display screen 630 is displayed. . The FAQ display screen 630 is, for example, a display screen as shown in FIG. 36, and a question display section 631, a related vocabulary index display section 632, an answer display section 633, and a "Return to FAQ candidate display screen" button 634 are displayed. ing. Here, when the user clicks the “Return to FAQ candidate display screen” button 634, the display on the display of the conversation control terminal 2 ′ returns to the FAQ candidate display screen 610.

  In the FAQ candidate display screen 610, when the user selects one of the related-word indexes displayed in the related-word index display unit 611 by clicking the mouse (arrow (2)), the related-word / co-occurring word list is displayed Screen 650 is displayed. The related word / co-occurrence word list display screen 650 is a display screen as shown in FIG. 37, for example, and the NO display part 651, the related word display part 652, the neighborhood related word display part (653 to 656), and the “FAQ "Return to candidate display screen" button 657 is displayed. Here, when the user clicks the “Return to FAQ candidate display screen” button 657, the display on the display of the conversation control terminal device 2 ′ returns to the FAQ candidate display screen 610.

  When the user selects one of the related words displayed in the related word display area 652 or the neighborhood related word display area (653 to 656) on the related word / co-occurrence word list display screen 650 by clicking the mouse, etc., An FAQ search screen 660 is displayed. The FAQ search screen 660 is, for example, a display screen as shown in FIG. 38B, which is substantially the same as the FAQ search screen 600 shown in FIG. It shows that it has returned to. Then, in the search keyword input portion 661 of the FAQ search screen 660, the related words (for example, “SNS” in the example of FIG. 37) selected on the related word / co-occurrence word list display screen 650 as the search keywords are automatically Set.

  In this situation, when the user clicks the “FAQ search” button 662, the FAQ candidate display screen 610 is displayed again, and a question sentence regarding “SNS” is displayed on the candidate question sentence display unit 612 this time.

<< Description about display processing of FAQ candidate display screen >>
Next, the display processing of the FAQ candidate display screen will be described with reference to FIG. FIG. 24 is a flowchart showing the display processing of the FAQ candidate display screen, and shows what processing is performed in the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′. In the conversation control terminal device 2 ′ ′, for example, each process is performed by the above-mentioned Topic 20, and the screen display on the display of the conversation control terminal device 2 ′ ′ shown in FIG. 35 to FIG. Or, it is performed by the WEB browser etc. which operate | moves by control of Topict20.

  First, in step S11, it is determined whether the user has clicked the “FAQ search” button 602 on the FAQ search screen 600. While the “FAQ search” button 602 is not clicked (NO), this determination is repeated. When the “FAQ search” button 602 is clicked (YES), in step S12, the search result by the search keyword input to the search keyword input unit 601 by the user is transmitted to the topic providing server 4 ′ as input information. In this embodiment, the search result is the result of performing a keyword search on a general Internet search site, and Topic 20 controls the search on this Internet search site, and sends the search result to the topic providing server 4 by API transmission. Send to '. The search result is, for example, the address of the web page that has hit the keyword search.

  When the topic providing server 4 ′ receives the input information from the conversation control terminal 2 ′ ′, in step S13, the input information is analyzed, the address of the web page included in the input information is accessed, and the HTML corresponding to the web page is accessed. An external log 502, which is the target text data, is acquired from data or the like.

  Next, in step S14, the topic providing server 4 'executes a sentence analysis process on the acquired external log 502, and extracts related words from the external log 502. The sentence analysis process will be described in detail later.

  Thereafter, in step S15, the topic providing server 4 'generates the related word dictionary 50 from the related words extracted from the external log 502 in step S14. The related-word dictionary 50 includes a related-word index 50 a for each piece of sentence information in the external log 502.

  Next, in step S16, the topic providing server 4 ′ acquires the related-word index 50a and the like from the related-word dictionary 50 in order to display it on the FAQ candidate display screen 610, and the conversation control terminal uses these information as input specific information. Send to device 2 ''.

  When the conversation control terminal device 2 ′ ′ receives the input identification information from the topic providing server 4 ′ (step S 17), the conversation control terminal device 2 ′ determines response information based on the received input identification information and the scenario data 28 in step S 18. The topic providing server 4 ′ transmits scenario data 55 to the conversation control terminal device 2 ′ ′ as necessary, and the conversation control terminal device 2 ′ ′ stores this in the scenario data 28.

  Next, in step S19, the response information determined in step S18 is displayed on the display of the conversation control terminal device 2 ''. For example, a FAQ candidate display screen 610 as shown in FIG. 35 (B) is displayed. In this embodiment, for example, a part of the collected question sentences (Q1, Q8, Q13, Q24, Q25) is displayed as a list of candidate question sentences on the candidate question sentence display unit 612. Further, in the related-word index display portion 611, related-word indexes respectively corresponding to the question sentences displayed as candidate question sentences are shown.

<< detailed explanation of sentence analysis processing >>
Next, with reference to FIG. 25, an outline of the sentence analysis process executed by the sentence analysis unit 43 (see FIGS. 20 and 21) of the topic providing server 4 ′ will be described. First, the sentence analysis unit 43 acquires the external log 502 which is text data (text data acquisition processing 520). External log 502 may be received from various data sources, as described above. In this embodiment, each web page is accessed based on the address of the web page received from the conversation control terminal 2 ′ ′, and text data is acquired from the corresponding HTML data or the like. When acquiring the acquired external log 502 or the external log 502, filtering processing may be performed to acquire only specific text data, or grouping may be performed according to a specific classification.

  Next, the sentence analysis unit 43 searches for the same (common) character string from the external log 502 acquired by the text data acquisition processing 520 (character string search processing 530). This process is, for example, a process of searching for and retrieving the same character string "Iroha" in the acquired external log 502. If 100 text strings "Iroha" exist in one text data, all of them are extracted. In addition, in the text data, there may be a plurality of the same character strings other than “Iroha”, but in that case, those character strings are also searched and extracted in the same manner. For example, in addition to the character string "Iroha", if there is a plurality of character strings "Nihone", the character string is extracted similarly. In addition, even when there are multiple character strings including "Iroha" which has already been taken out as the same character string, such as the character string "Iroha", the same character string is searched separately from the character string "Iroha" Be done.

  The character string search process 530 further stores the character string searched as the same character string in the search result data 48 together with the preceding character and the following adjacent character of the character string. The data stored in the search result data 48 is, for example, for the character string "Iroha" in the above example, for each of the 100 "Iroha", adjacent characters before the character strings "Iroha" and "Iroha", And it is data including the adjacent character after "Iroha". In the case of the above-described example, the character string, the preceding adjacent character, and the subsequent adjacent character are similarly stored in the search result data 48 for the character string “to be soft” and the character string “to be”.

  Thus, when the same character string is searched in the character string search process 530, storing the character string and the adjacent characters before and after it ultimately determines the relevancy that has an important meaning. However, even if the same string exists in the external log 502, if the appearance frequency does not reach a predetermined frequency, it may be determined as a related word at this time. It can be determined that the data is not stored as the search result data 48 regarding the character string. This is because it is possible to determine that a word (character string) that appears only a few times in the external log 502 consisting of many characters is not of high importance in the first place.

  Further, in the character string search process 530, in order to search for and store the same character string from the external log 502, in the present embodiment, a search data structure called a suffix array is used and divided into two. By searching by Binary Search, the same string is searched at high speed. In the present embodiment, the character string search processing 530 is performed by the above method, but various other methods can be adopted to perform the same search processing. The processing of the character string search processing 530 using the suffix array and the binary search will be described in detail later.

  Next, the sentence analysis unit 43 determines from the character string stored in the search result data 48 by the character string search processing 530 and the contents of the adjacent characters before and after it, the degree of difference between the previous adjacent character and the degree of difference between the subsequent adjacent characters. Are determined (difference degree determination processing 540).

Here, one character is represented as s (i),
A character string m (i, j) composed of s (i) to s (j),
m (i, j) = (s (i), s (i + 1), s (i + 2),... s (j-2), s (j-1), s (j)),
A character string m (i, j-1) composed of s (i) to s (j-1) is
m (i, j-1) = (s (i), s (i + 1), s (i + 2), ... s (j-2), s (j-1)),
A character string m (i + 1, j) composed of s (i + 1) to s (j) is
m (i + 1, j) = (s (i + 1), s (i + 2),..., s (j-2), s (j-1), s (j)).

In this case, the boundary condition for the previous adjacent character is
T (i-1) = {S (i-1) | m (i, j)}
T (i) = {S (i) | m (i + 1, j)}
Defined by
The boundary conditions for the later adjacent characters are
B (j) = {S (j) | m (i, j-1)}
B (j + 1) = {S (j + 1) | m (i, j)}
Defined by

  Here, for example, {S (i-1) | m (i, j)} makes the character string m (i, j) common, and means a set of characters appearing immediately before that. Here, s (i) ∈T (i) and s (j) ∈B (j) hold.

  For example, if the number of elements of T (i-1) is large and the number of elements of T (i) is 1, s (i) is likely to be the beginning of the related words, while elements of B (j) When the number is 1 and the number of elements in B (j + 1) is large, s (j) is likely to be the end of the related words, and as a result, the string m (i, j) is a candidate of related words It is judged as

  Thus, for the same character string, the degree of difference regarding the adjacent character for the same character string is determined based on the appearance of the adjacent characters before (or after) those characters, that is, how many variations the adjacent character appears. judge. When the degree of difference regarding the adjacent characters before and after is determined, the determined degree of difference is stored in the related word candidate data 49 together with the corresponding character string. The process of the difference degree determination process 540 will be described in detail later.

  Next, the sentence analysis unit 43 determines whether or not the same character string is a related word, based on the degree of difference regarding the adjacent characters before and after the same character string determined by the difference degree determination processing 540, If it is determined to be a related word, the character string is stored in the related word dictionary 50 (related word determination process 550).

  As mentioned above, if you look at what kind of variation there is in the adjacent characters before and after the same character string, if the number of appearance elements of the adjacent character is small, the adjacent character and the "same character string" Can be considered to form another commonly used character string, while if the number of appearance elements of the adjacent character is large, the “same character string” is separated from the adjacent character, and the “same character It is highly probable that the column is an independent term and a high importance term. Further, whether or not the “same character string” is a related word can be determined in consideration of additional factors in addition to the degree of difference of the preceding adjacent character and the degree of difference of the subsequent adjacent character. The related term determination process 550 will be described in detail later.

  Furthermore, in the related word determination process 550, when a plurality of related words are determined, it is possible to perform ranking among the character strings determined as the related words. Such a ranking is, for example, a ranking with respect to the importance of a string, and can be determined in consideration of additional factors in addition to the degree of difference between adjacent characters before and after the string. For example, ranking can be performed based on the character length of the character string, the appearance frequency, and the like. In addition to showing the rankings, it is also possible to rank them numerically so that they can represent relative degree differences.

<Explanation of sentence analysis process for specific external log example>
FIG. 26 shows an external log 502 a which is an example of the external log 502. The external log 502a is, as shown in FIG. 26A, a collection of only sentence information including "network" in the described part of the question sentence as a search result by the search keyword, and the original data is, for example, , Text data for solving problems written by various users on various servers on the Internet. Typical examples of such text data include web pages (home pages) and text data of blogs published on the Internet, and tweet information of TWITTER. In addition, data generated and edited in advance by any institution or text information in a database may be included. In the original data, it is assumed that a question sentence (Q (question)) and an answer sentence (A (answer)) are configured to correspond one to one. Here, only the sentence information in which the character string “network” is included in the written part of the question sentence is extracted from the original data, but the character string “network” is included in the written part of the question sentence and the answer sentence Various variations can be considered such as extracting sentence information.

  Here, when the sentence analysis process is performed by the sentence analysis unit 43, the external log 502a shown in FIG. 26A is acquired by the text data acquisition process 520, and thereafter, the above-described character string search process 530, difference degree determination When the process 540 and the related word determination process 550 are performed, as shown in FIG. 26B, a plurality of related words are extracted for the question sentence of the extracted sentence information. For example, among the question sentences, “network”, “trouble”, “response”, and “setting” are selected for Q1. In the description of the question in Q1, the "network" corresponding to the search keyword is underlined, and the other related words are enclosed by a rectangle. Also, {network, setting, trouble, correspondence} is shown as a set of extracted related words corresponding to the description of the question in Q1. The set of related words is described in the order of ranking described above with reference to the related word determination process 550.

  Similarly, for Q8, "network", "setting" and "by event" are selected. In the description of the question in Q8, the "network" corresponding to the search keyword is underlined, and the other related words are enclosed in a rectangle. In addition, {network, setting, by event} is shown as a set of extracted related words corresponding to the description of the question in Q8. Further, it is also possible to extract relevant words not only for the question sentences but also for the description contents of the answer sentences and include them in the related vocabulary index.

  An example of the related word dictionary 50 generated by such a related word determination process 550 is shown in FIG. 27 (A). A set of related words related to each question shown in FIG. 26B is stored as the related word index 50a as it is. Although it may be sufficient to store only the related-word index 50a as the related-word dictionary 50, in this embodiment, the question sentence corresponding to the related-word index is stored in the question sentence 50b, and The answer corresponding to the question sentence 50b is stored in the answer sentence 50c.

  As shown in FIG. 27A, the related-word index 50a stored in the related-word dictionary 50 is stored in association with corresponding sentence information, whereby the set of one related word and the other related words are stored. Related words belonging to other sets are associated with each other through related words common to the set of. For example, as shown in FIG. 27B, {network, setting, trouble, correspondence} is grasped as a set of related words for the question Q1 among the question sentences, and the question Q8 among the question sentences If {network, setting, by event} is grasped as a set of related words of, there is a related word “setting” which is common (except for the related word “network” which is a search keyword), , The relative word "setting" and the related word related to the term "co-occurrence related word" can be understood to be {by trouble, correspondence, separately by event}, and the related words "trouble" and "correspondence" are further related It is understood that they have a relationship in that they are related to the word "by event", that is, they both have a common point of co-occurring with the related word "setting".

  Such a relation between related words is found between the sentence information of the external log 502a collected by the search keyword "network" in this embodiment, but is collected by a completely different search keyword. In some cases, it is possible to find such a relation between related words of sentence information, and in this case, it is also possible to find a relation of potential topics.

<Detailed Description of Character String Search Processing in Sentence Analysis Unit>
The character string search process 530 will be described with reference to FIGS. 28 to 30. FIG. 28 is a flowchart showing the procedure of the character string search process 530. FIGS. 29 and 30 are diagrams showing a character string search mechanism using a suffix array and a binary search, and for example, a part of the character string of the external log 502 is used as the character string to be searched. Text data 502-1 of “in the code list” is set. Normally, all the text part of the external log 502 is to be searched, but here, for the sake of illustration, it is part of the text part.

  First, a suffix array is created to search for the same string in the string "this code is in the code list". In step S21 of FIG. 28, suffixes are expanded from the first character of the text data to the last character. Here, referring to FIG. 29, as shown in FIG. 29 (B), the text data 502-1 to be searched (“this code is in the code list”) shown in FIG. 29 (A). The suffixes indexed from 1 to 15 are expanded. Each suffix is a character string from the position of the index (start character position) to the end in the text data to be searched 502-1, and for example, for the index "1", the text data to be searched is 502-1. A string from the first character to the end (15th character) "This code is in the code list" is shown. Further, for the index “10”, a character string “in the list” from the 10th character to the end (15th character) of the text data 502-1 to be searched is shown. For the last index "15", the character "Ru" at the end of the 15th character is shown.

  Next, in step S22 of FIG. 28, the expanded suffixes are sorted in a predetermined order to create a suffix array. Here, referring to FIG. 29, the suffixes shown in FIG. 29 (B) are sorted, and the suffix array after sorting is shown in FIG. 29 (C). Sorting is performed, for example, by a character code (JIS code in this example) corresponding to each character, and from the first record to the fourth record, the first character is "-" (JIS code = 213C), "A" As shown in (JIS code = 2422) and “GA” (JIS code = 242C), it is understood that the character codes are sorted in the order of the character codes. The first and second records have the same first and second characters, but the third character is different in "ga" and "ri" ("ga" (JIS code = 242C), "ri" ( JIS code = 256A)), whereby the first record is placed first. The index shown in FIG. 29B is attached as it is to the corresponding record.

  Next, in step S23 of FIG. 28, one search character string is sequentially determined from the text data. Since this is to search for the same string in the text data, all the substrings in the text data are matched with the text data as a search string. For example, with regard to the search target text data 502-1 in FIG. 29, the search character strings “K” to “R” of one character, the search character strings “this”, “ko” of two characters,. "Aru", "Aru", three-character search character string "this ko", "no ko", ..., "To nia", "Aru", etc. are decided in order, and so on, and so on. "This code is in the code list" is the search string. However, in this case, the search character string of one character or 15 characters can be omitted because it does not have meaning to be compared with the text data 502-1 to be searched. In addition to that, the character length of the search character string can be limited to a predetermined range.

  Next, in step S24 of FIG. 28, it is determined whether all search character strings to be searched have ended. Here, when all processing is completed, the character string search processing is completed. If all have not been completed yet, in step S25, the suffix string is searched using the search string as a search key. Here, referring to FIG. 30, as shown in FIG. 30 (A) and FIG. 30 (B), the character string "code" of three characters is determined from the search target text data 502-1 as the search character string. The search process is shown when it is done. At this time, a binary search is performed on the suffix array of FIG. The suffix array of FIG. 30 (C) is the same as the suffix array shown in FIG. 29 (C), and FIG. 30 (C) shows the process of search by binary search.

  Referring to FIG. 30C, first, a record (1) (index = 15, character string “R”) at the center of the suffix array is compared with the search character string “code”. At this time, the JIS code of “ru” is 246 B, and the JIS code of “ko” is 2533, and “ko” is larger, so the record located at the center of the lower half of the suffix array below record (1) 2) Compare the search string "code" with (index = 12, the string "is in"). At this time, the JIS code of "to" is 2548, and the JIS code of "ko" is 2533. Since "ko" is smaller, record (3) (3) (2) located at the center of record (1) and record (2) Index = 7, compare the string "in code list") with the search string "code".

  Then, since the first three characters of the record (3) match the search character string “code”, the same character string as the search character string is found in the search target text data 502-1. After that, the first three characters of the record (4) match the search character string “code” when compared with the upper and lower records of the record (3), so the text data 502 that is the same as the search character string is searched It means that -1 has been found. By such binary search, it can be seen that there are two character strings "code" in the text data 502-1 to be searched (one of them is extracted from the text data 502-1 to be searched) Search string "code" is itself).

  Next, in step S26 of FIG. 28, it is determined whether the search character string has hit a predetermined number. If it is determined that the search character string does not hit a predetermined number, the process proceeds to step S23 in order to perform a search with the next search character string without using the character string as a candidate of a related word. Here, the predetermined number may be determined based on various factors such as the number of characters of text data 502-1 to be searched and the number of characters of the search character string. The absence of a predetermined number of hits means that the appearance frequency of the search character string is small in the search target text data 502-1, and it is not an important word. At this stage, without evaluating the appearance frequency, it may be stored as a candidate of related words and may be finally determined in the subsequent related word determination process or the like.

  If it is determined in step S26 that the search character string has hit a predetermined number, the process proceeds to step S27, where the character string matching the search key (search character string) is regarded as a candidate of a related word, and each adjacent character before and after And stored in the related word candidate data 49 as one record. Here, referring to FIG. 30C, as a search result, for the record (3) and the record (4) in which the character string matching the search character string “code” is found at the beginning, the same as the search character string The character string "code" and the adjacent characters before and after are stored as one record. For example, for the record (3), "ga", "code" and "ri" are stored as one record. The previous adjacent character is "ga" and the subsequent adjacent character is "ri". Also, for the record (4), "no", "code" and "ga" are stored as one record. The previous adjacent character is "no", and the subsequent adjacent character is "ga".

  As described above, as described above, the sentence analysis unit 43 in this embodiment is configured to quickly find the same character string in the text data using the suffix array and the binary search, but the present invention It is not limited to the said processing method. It is possible to find the same string in the text data in a manner other than that described above.

<Detailed Description of Difference Degree Determination Processing in Sentence Analysis Unit>
Next, with reference to FIGS. 31 and 32, the difference degree determination process of the sentence analysis unit 43 in this embodiment will be described in more detail.

  FIG. 31 is a flowchart showing the procedure of the difference degree determination process 540. FIG. 32 is a diagram showing a mechanism for determining the degree of difference between adjacent characters before and after. A character string search process with a search character string “code” as shown in FIG. 30 is a search including many characters As a result of performing the search for the target character string, 26 search character strings “code” are obtained as the search result, and the situation where the corresponding 26 records are processed is shown.

  First, in step S31 in FIG. 31, in the character string search process 530, a record relating to one character string is extracted from the character string records (including adjacent characters before and after) stored in the search result data 48. Here, referring to FIG. 32A, it is shown that a record (26 records in all) regarding the character string “code” stored in the search result data 48 is extracted and expanded in the memory.

  Next, in step S32 of FIG. 31, all records of the character string stored in the search result data 48 are acquired, and when it is determined that there is no data to be subjected to difference degree determination processing, the difference degree determination processing of FIG. Ends. If it is determined in step S32 that all the processes have not been completed and all the records for one of the character strings stored in the search result data 48 have been acquired, the process proceeds to step S33.

  Next, in step S33 in FIG. 31, all records acquired for one of the character strings stored in the search result data 48 are sorted by the previous adjacent character, and the number of patterns of appearing characters for the previous adjacent character is calculated. Ask. Here, referring to FIG. 32A, results of sorting the records acquired for the character string “code” stored in the search result data 48 (total 26 cases) by the previous adjacent character 561 are shown. . This sort can be performed using a character code (for example, JIS code) as in the case of the creation of the suffix array in the character string search process described above. After performing such sorting, the number of patterns of appearance characters of the previous adjacent character 561 can be known by counting the number of times the value of the previous adjacent character 561 changes (breaks) between records. In the case of FIG. 32A, the previous adjacent character 561 is “”, “(”, “,”, ““, “が”, “た”, “ta”, “de”, “do”, “no”, There are 13 patterns of "ha", "be", "mo" and "ri".

  Thus, in the sentence analysis unit 43 of this embodiment, the previous adjacent character 561 is sorted, and the number of patterns of appearing characters related to the previous adjacent character is obtained based on whether or not the value has changed between each record. However, the present invention is not limited to the treatment method. The number of patterns of appearing characters can be determined in various other ways. Further, the handling of a predetermined character or character pattern can be flexibly set according to the specification. For example, control characters such as line feed and page feed can be not considered. Also, it is possible to take into consideration or ignore punctuation marks and reading points. Furthermore, upper and lower case letters can be treated as the same letter or as different letters. Further, for sorting, for 1-byte code characters such as half-width alphabets and numbers, the corresponding 1 byte is sorted, and for 2-byte code characters such as Kanji, the corresponding 2 bytes are sorted. Also, in the case of sorting the previous adjacent characters 561, in the present embodiment, it is assumed that one adjacent character is sorted, but characters of two or more characters can be sorted to determine the degree of difference between them. .

  Next, in step S34 in FIG. 31, the degree of difference regarding the previous adjacent character 561 is determined. In this case, the degree of difference is an index indicating how much the adjacent character 561 differs (between the 26 records in FIG. 32A), and therefore, the appearance of the preceding adjacent characters described above It is determined based on the number of patterns. The degree of difference may be the pattern number itself, but it can also be determined in consideration of, for example, the appearance frequency of the character string (in the case of FIG. 32A, the appearance frequency of “code” is 26), etc. . Also, the degree of difference can be evaluated in a plurality of stages (for example, three stages) by a predetermined threshold. Furthermore, in the case where the adjacent character 561 is a specific character, arbitrary weighting can be performed on the count of the character and the determination of the degree of difference.

  Next, in step S35 in FIG. 31, all records acquired for one of the character strings stored in the search result data 48 are sorted by the subsequent adjacent characters, and the number of patterns of appearing characters for the subsequent adjacent characters is calculated. Ask. Here, referring to FIG. 32 (B), there is shown the result of sorting the records (26 in all) acquired for the character string “code” stored in the search result data 48 by the adjacent character 563 later. . As a result of such sorting, the records 565 and 566 shown in FIG. 32A are arranged at the positions indicated by the dotted arrows.

  This sort can be performed using a character code (for example, JIS code) as in the case of the creation of the suffix array in the character string search process described above. After performing such sorting, the number of patterns of appearance characters of the subsequent adjacent character 563 can be known by counting the number of times the value of the subsequent adjacent character 563 has changed (broken) between records. In the case of FIG. 32B, the subsequent adjacent characters 213 are "," "," "," "," "," "," "," "," "," "," "," There are 12 patterns of “H”, “R” and “P”.

  Thus, in the sentence analysis unit 43 of this embodiment, the subsequent adjacent characters 563 are sorted, and the number of patterns of appearing characters related to the subsequent adjacent characters is calculated based on whether the value has changed between each record. However, the present invention is not limited to the treatment method. The number of patterns of appearing characters can be determined in various other ways. Also, predetermined characters may not be included in the count of the number of patterns. Further, for sorting, for 1-byte code characters such as half-width alphabets and numbers, the corresponding 1 byte is sorted, and for 2-byte code characters such as Kanji, the corresponding 2 bytes are sorted. Further, in the case of sorting the adjacent characters 563 later, in the present embodiment, the adjacent characters of one character are sorted, but it is possible to sort the characters of two or more characters and determine the degree of difference thereof. .

  Next, in step S36 in FIG. 31, the degree of difference regarding the subsequent adjacent character 563 is determined. In this case, the degree of difference is an index showing how much the adjacent character 563 differs (between the 26 records in FIG. 32B), and therefore, the appearance of the adjacent character after the above It is determined based on the number of patterns. The degree of difference may be the pattern number itself, but it can also be determined in consideration of, for example, the appearance frequency of the character string (in the case of FIG. 32B, the appearance frequency of “code” is 26), etc. . Also, the degree of difference can be evaluated in a plurality of stages (for example, three stages) by a predetermined threshold. Furthermore, in the case where the adjacent character 563 is a specific character, arbitrary weighting can be performed on the count of the character and the determination of the degree of difference.

  Next, in step S37 in FIG. 31, the character string to be determined and the degree of difference with respect to the previous adjacent character and the degree of difference with respect to the subsequent adjacent character, which are determined with respect to the character string, are stored in the related word candidate data 49 Be done.

  When the process of step S37 of FIG. 31 ends, the process proceeds to step S31, and the process regarding the next "same character string" is performed.

<Detailed Description of Related Word Determination Process in Sentence Analysis Unit>
In the related word determination process 550, the data stored in the related word candidate data 49 is sequentially read out by the difference degree determination process 540, and the corresponding character string is a related word based on the determined difference between adjacent characters before and after It is determined whether or not it is determined, and the determined related words are stored, for example, in the related word dictionary 50 in the form of the related word index described above, and sentence information corresponding to the related word index (for example, question in the above example) The sentence and the answer sentence are stored in the related word dictionary 50. The related word determination processing 550 determines, for example, whether or not the corresponding character string is a related word, according to the magnitude of the degree of difference between adjacent characters before and after.

  The degree of difference between the adjacent characters before and after can be determined by a common or different judgment criterion before and after. When the degree of difference between adjacent characters before and after is determined to be a predetermined size, the corresponding character string is determined as a related term which is an independent term and is an important word for identifying a topic. . That is, a score may be calculated based on the degree of difference between adjacent characters before and after, and whether or not the corresponding character string is a related word may be determined based on the score.

  In addition, in the related word determination process 550, in addition to the degree of difference between adjacent characters before and after, the character length of the corresponding character string, the appearance frequency of the corresponding character string, the probability and frequency that a specific character appears as the previous adjacent character The score is calculated based on the calculated score, taking into consideration the probability / frequency that a particular character appears as a subsequent adjacent character, and the probability / frequency that a particular character combination appears as an adjacent character before and after, etc. It may be determined whether the corresponding character string is a related word.

  Furthermore, when there is a plurality of candidates for related words in one text data (character string to be searched), it is determined as a related word according to the number of candidates and the distribution of scores calculated for each of the candidates. It is also possible to change the decision conditions of.

  Furthermore, in the related word determination process 550, when there are a plurality of related words determined for each of the external log 502 or the sentence information included in the external log 502, the strings determined as related words are ranked. Can be done. Such a ranking is, for example, a ranking regarding the importance of a character string related to a topic, and in addition to the score calculated by the degree of difference between adjacent characters before and after or the degree of difference between adjacent characters before and after These factors are determined based on the score etc. calculated in consideration of other factors. Also, such ranking can not only order the importance of the character string determined as a related word but also indicate the relative importance between related words, for example, the above-mentioned score It is possible to rank by a concrete numerical value by using a value or the like.

  In addition, although such ranking is performed when a plurality of related words are determined, a plurality of related words determined for one text data can also be performed, or a plurality of groups grouped according to a predetermined condition It can also be performed on a plurality of related words determined for text data.

  For example, as a plurality of text data grouped under a predetermined condition, text data of a web page group that hits the input search keyword, the content of the TWITTER of the user corresponding to the predetermined attribute, and the like can be considered.

<< Description about display processing of FAQ display screen >>
Next, the display processing of the FAQ display screen will be described with reference to FIG. FIG. 33 is a flowchart showing the display processing of the FAQ display screen, and shows what processing is performed in the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′. In the conversation control terminal 2 ′ ′, for example, each process is performed by the above-described Topic 20.

  First, in step S41, the user selects one of the candidate question sentences displayed in candidate question sentence display portion 612 on the FAQ candidate display screen 610 shown in FIG. It is determined whether or not it is. While one of the candidate question sentences is not selected (NO), this determination is repeated. If one of the candidate question sentences is selected (YES), the selected candidate question sentence is transmitted to the topic providing server 4 'as input information in step S42. The input information may include the selected question itself, but it is sufficient if it includes an identifier that can identify this question. The user can also select a plurality of question sentences of interest simultaneously.

  When the topic providing server 4 ′ receives the input information from the conversation control terminal device 2 ′ ′, in step S43, the input information is analyzed, and the answer sentence corresponding to the question sentence included in the input information is acquired from the related word dictionary 50 Do. In this embodiment, as shown in FIG. 27A, the related sentence dictionary 50 stores the question sentence 50b and the answer sentence 50c corresponding thereto. However, these question sentences 50b and the answers are stored. The sentence 50c can be stored in another file while being associated with the relevancy index 50a of the relevancy dictionary 50.

  Next, in step S44, the topic providing server 4 'causes the information including the answer sentence 50c corresponding to the question sentence 50b acquired from the related word dictionary 50 to be displayed in the FAQ display screen 630 as the related word / co-occurrence word While storing in the data 52, this information is transmitted as input specifying information to the conversation control terminal device 2 ''.

  When the conversation control terminal device 2 ′ ′ receives the input identification information from the topic providing server 4 ′ (step S 45), the conversation control terminal device 2 ′ determines response information based on the received input identification information and the scenario data 28 in step S 46. The topic providing server 4 ′ transmits scenario data 55 to the conversation control terminal device 2 ′ ′ as necessary, and the conversation control terminal device 2 ′ ′ stores this in the scenario data 28.

  Next, in step S47, the response information determined in step S46 is displayed on the display of the conversation control terminal device 2 ''. For example, in the FAQ candidate display screen 610 as shown in FIG. 35B, one of the question sentences listed in the candidate question sentence display portion 612 (for example, the question of Q24 shown by the arrow (1)) is selected. Then, the FAQ display screen 630 as shown in FIG. 36 is displayed, where the question display section 631 displays the question of Q24 which is the selected question sentence, and the related vocabulary index display section 632, Q24's. The related-word index corresponding to the question is displayed, and further, the answer display unit 633 displays an answer (answer of A24) corresponding to the question of Q24.

  By such screen transition from the FAQ search screen 600 to the FAQ display screen 630, the user can search the FAQ with the search keyword designated by the user, and can display a plurality of question sentence candidates as a search result. Furthermore, there is a need to easily grasp what important keywords appear in each candidate question sentence (that is, what matters are related to them) by looking at the related-word index. it can.

  Further, in this embodiment, although the answer sentences corresponding to the question sentences are not displayed on the FAQ candidate display screen 610, the corresponding answer sentences should be displayed at the stage of displaying the candidate question sentences as a list. You can also.

<< Description about display processing of related words / co-occurrence word list screen >>
Next, referring to FIG. 34, display processing of the related word / co-occurrence word list screen will be described. FIG. 34 is a flowchart showing display processing of a related word / co-occurrence word list screen, and shows what kind of processing is performed in the conversation control terminal device 2 ′ ′ and the topic providing server 4 ′. In the conversation control terminal 2 ′ ′, for example, each process is performed by the above-described Topic 20.

  First, in step S51, the user selects one of the related-word indexes displayed on the related-word index display unit 611 on the FAQ candidate display screen 610 shown in FIG. It is determined whether or not it is. This determination is repeated as long as one of the related index is not selected (NO). If one of the related index is selected (YES), the selected related index is transmitted to the topic providing server 4 'as input information in step S52. The input information may also include the selected related-word index itself, but it is sufficient if it includes an identifier that can identify this related-word index. Note that the user can simultaneously select a plurality of related word indexes of interest.

  When the topic providing server 4 'receives the input information from the conversation control terminal device 2' ', in step S53, the input information is analyzed, and all the related words including the related-word index included in the input information from the related word dictionary 50 Get index

  Next, in step S54, the topic providing server 4 'performs preference analysis on all the related words included in the related word index acquired in step S53 based on the internal log 506 such as the preference data 51. The preference data 51 is a log file storing data indicating usage patterns such as how the related words have been used so far by each user, as shown in FIG. 38A, for example. In addition, for each user ID 51a, it is stored which related word 51b the user has input as a reference or search keyword. Besides, it is also possible to store usage date and time of relative words and detailed usage contents as preference data 51, and to perform preference analysis taking these information into consideration as well.

  The topic providing server 4 ′ performs preference analysis on all the related words included in the conjunction index, based on the preference data 51, and determines the degree of importance of each related word. For example, with reference to the preference data 51, in the same user, the relevance of the related words is set such that the higher the frequency of use of the related words, the higher the importance. The related words associated by such importance levels correspond to the topic key (cluster) 504 described above.

  Next, in step S55, the topic providing server 4 'performs topic analysis on the related words whose importance is set in step S54 based on the topic data 507 such as the topic data 53. The topic data 53 may be a topic input by a contractor's person in charge, or a topic automatically extracted from the external log 502 based on a topic input by the person in charge. Based on such topic data 53, the distribution is captured, and the distribution of related words associated with each other is provided to the user. For example, it is possible to associate a related word that is a topic in the FAQ with its co-occurring word so as to indicate how the related word is distributed in the topic. Further, in the input and the automatic extraction of the topic material, the adjustment can be performed for each user of the conversation control terminal device 2 ′ ′, and the related words can be recommended according to the user. For example, the topic can be adjusted to limit the target question sentence to a predetermined range.

  Next, in step S56, the topic providing server 4 'uses the information including the associated words finally associated in step S55 as the input specifying information to be displayed on the associated word / co-occurring word list display screen 650. Transmit to the conversation control terminal 2 ''.

  When the conversation control terminal device 2 ′ ′ receives the input identification information from the topic providing server 4 ′ (step S 57), it determines the response information based on the received input identification information and the scenario data 28 in step S 58. The topic providing server 4 ′ transmits scenario data 55 to the conversation control terminal device 2 ′ ′ as necessary, and the conversation control terminal device 2 ′ ′ stores this in the scenario data 28.

  Next, in step S59, the response information determined in step S58 is displayed on the display of the conversation control terminal device 2 ''. For example, a related word / co-occurrence word list display screen 650 as shown in FIG. 37 is displayed on the display of the conversation control terminal device 2 ''. In the related word / co-occurrence word list display screen 650, an NO display part 651, a related word display part 652, a neighborhood related word display part (653 to 656), and a "Return to FAQ candidate display screen" button 657 are displayed. Related words are displayed in the form of a two-dimensional matrix.

  With respect to the vertical axis direction of the matrix, in the related word display section 652, all related words appearing in the related word index of the related word dictionary 50, which are extracted as related words for the FAQ search of this embodiment, are displayed in order without duplication. There is. The display order is set according to the importance of each relevance determined by the preference analysis, and in this embodiment, the lower the number shown in the NO display portion 651 (the more it is displayed in the upper part of FIG. 37) , The relevant relevancy is high in importance. In FIG. 37, the numbers shown in the NO display portion 651 are from 1 to 17. However, operating the slider bar of the related word / co-occurrence word list display screen 650 to move downward Allows you to browse more related words.

  In relation to the horizontal axis direction of the matrix, on the right side of the related words displayed in the related word display area 652, the neighborhood related word display area 653 to the neighborhood related word display area 656 are shown. The neighborhood related words related to the related words displayed in are displayed. In the vicinity, as described with reference to FIG. 16, when focusing on a certain related word A, the related word set including the related word A is “near the related word”, and such related words are referred to as the near here. It is considered as a related word. Note that the related word set is a set of related words included in a certain topic, and here, the related word index corresponds to this.

  In FIG. 37, although only four neighborhood relevancy words from neighborhood relevancy words 1 to 4 are displayed, the slider bar of the relevancy word / co-occurrence word list display screen 650 is operated in the right direction. By moving to, you can browse more neighborhood related words. The vicinity related words can be said to be related words that co-occur with the related words displayed in the related word display portion 652 (co-occurrence related words: that is, related words that appear together in the same topic). The display order (in the horizontal direction) of the neighborhood related words is displayed in the related word display area 652 as the co-occurrence relation is strong, that is, the one that appears more frequently with the related word displayed in the related word display area 652. Are adjusted to be displayed near the related words. The display order in the horizontal direction can also be determined in consideration of ranking of related words, setting by a user or an information search system, and the like.

  Further, in the related word / co-occurrence word list display screen 650, among the related words displayed in the related word display portion 652, the related word index display portion 611 is displayed on the FAQ candidate display screen 610 shown in FIG. When one of selected related index is selected by mouse click or the like, related items included in the related index (in this example, for example, “network”, “connection”, “router” And “modem” are highlighted for easy recognition by the user.

  The user originally focused on the matrix display of the related words displayed on the related word display portion 652 of the related word / co-occurrence word list display screen 650 and the vicinity related word display portion (653). Seeing another related word related to a related word (search keyword) in terms of how it has used the related word and in terms of whether it is close to a topic to be focused or recommended by the user or the information search system It is possible to be given "awareness" about new related words. For example, in the matrix display of related words in the related word / co-occurrence word list display screen 650 of FIG. 37, a search is performed using the search keyword “network”, but the numbers displayed on the NO display portion 651 are 1 to 13, In the case of and 15, it is understood from the set of displayed related words that the topic is generally related to the communication network, but when the numbers displayed on the NO display portion 651 are 14, 16, and 17, It is a topic related to social networks, and it can be understood that a topic different from other topics has appeared.

  When the user selects one of the related words displayed on the related word display portion 652 of the related word / co-occurrence word list display screen 650 or the vicinity related word display portion (653) by clicking the mouse (FIG. 37) Arrow, and the screen automatically transits to the FAQ search screen 660 as shown in FIG. 38 (B), where the selected relevance (in this example, the “SNS” of the proximity relevance) is The search keyword input portion 661 of the FAQ search screen 660 is automatically set. In this situation, when the user clicks the “FAQ search” button 662, the FAQ candidate display screen 610 is displayed again, and a question sentence regarding “SNS” is displayed on the candidate question sentence display unit 612 this time.

<<< Description of outline of processing in information update unit >>>
The information updating unit 46 extracts a character string that can be semantically identified from the external log 502 (text data) collected according to different collection conditions, and stores the extracted character string in a related-word dictionary corresponding to the text data. The comparison results obtained by performing comparison processing on these related words dictionaries are stored and updated in the comparison result data 54. The comparison process is automatically performed when the relevancy dictionary is updated.

  The text data associated with different association dictionary is text data collected by different collection conditions, and these text data are, for example, a plurality of text data collected at different timings for similar objects or data sources. Or, at the same timing, it may be plural text data collected by different subjects or search conditions.

  The above comparison processing compares a plurality of related words dictionaries, and in addition to the newly appeared related words, the deleted related words, the commonly appeared related words, (corresponding to three or more time-series text data In the dictionary, the appearance condition of the relation words, such as the relation words appearing again, is determined, and if the relation words are any of such appearance situations, the relation words are stored as a comparison result.

  In addition, when a plurality of related words are extracted from one text data, those related words can be associated as one set (as co-occurrence related words) and stored in the corresponding related word dictionary. By configuring in this way, when it is determined in the comparison process that there is a related word common to a plurality of related word dictionaries, it is possible to compare co-related related words of the related words and perform further determination. .

  In this way, by capturing the history of the appearance of related words, it becomes possible to reveal the meaning of related words. That is, by repeating such processing, the topic name to which the relative term belongs becomes clear, and it is possible to distinguish between the regular related term and the newly-relevant related term (which has newly appeared), or the relative term By comparing the dictionaries, it is possible to judge the similarity and the difference of the topics. Such a function is referred to as a related-word learning function. The related-word learning function can be expected to diversify the input identification means of the end user.

  The process outline of the information updating unit 46 will be described with reference to FIG. First, the information updating unit 46 acquires the external log 502 which is text data (text data acquisition processing 700). The external log 502 is collected by the crawler 730, for example. When the crawler 730 returns the network address (URL etc.) of the WEB page, the external log 502 can be acquired by accessing the network address. Furthermore, when acquiring the acquired external log 502 or the external log 502, filtering processing may be performed so as to acquire only specific text data, or grouping may be performed according to a specific classification.

  The crawler 730, for example, performs topic analysis on topic names automatically started and decided at a determined time (ie, performs a search and periodically collects topics). The topic name is stored, for example, in the service holding the related word dictionary 50 (the area assigned to the service ID corresponding to the topic handled by the user), and if the user wants to handle 10 topics, 10 You will use these services to handle those topics. In addition, corresponding topic chips are set for each service described above, and each topic chip constantly collects information on the corresponding topic, and causes related topic chips to be linked and integrated according to the user's input. It is also possible to realize more diverse topic offering services.

  The search by the crawler 730 receives a search result from the search server of the said internet search site by, for example, accessing the existing internet search site on the internet and specifying a search keyword there. In the search result, for example, the address of the WEB page containing the content matching or similar to the search keyword (the address of the WEB page 1, the address of the WEB page 2, the address of the WEB page 3, ..., the address of the WEB page X Is included.

  In this embodiment, the crawler 730 is configured to acquire search results by executing a search on an existing Internet search site, but various other methods acquire an address of a web page that satisfies a predetermined condition. can do. Also, the search target is not limited to WEB pages on the Internet, and TWITTER's tweet information and data (on network or local) generated in advance by any organization or organization or text in a database It may be information.

  An existing Internet search site receives a search request from a data source on the Internet, from the data source on the Internet, using the search engine provided in the search server used by the Internet search site, the address of the web page matching or similar to the search keyword. Collect (or by prior periodic collection activities).

  When the search results are transmitted from the search server, the crawler 730 transmits the search results (for example, by API transmission) from the computer on which the crawler 730 operates to the topic providing server 4 ′. In addition, the crawler 730 can be configured to use filters to exclude search results that satisfy a predetermined condition.

  In this embodiment, the crawler 730 is automatically activated at a predetermined time, but may control the operation of the crawler 730 according to the operation control of the topic providing server 4 'to acquire the search result. Good. Also, the crawler 730 acquires the search results at predetermined intervals, holds the search results in the computer on which the crawler 730 operates, and the topic providing server 4 ′ accesses and searches the computer at the necessary timing. You can also get the result. Also, the crawler 730 can be configured to be executed by the topic providing server 4 '.

  Further, in this embodiment, the computer on which the crawler 730 operates transmits the address of the web page related to the search keyword as the search result to the topic providing server 4 ′. However, on the computer on which the crawler 730 operates, these web pages , And the resulting text data can be transmitted as the external log 502 to the topic providing server 4 ′.

  Next, the information updating unit 46 extracts a character string that can be semantically identified from the external log 502 acquired by the text data acquisition processing 700, and stores the extracted character string in the related word dictionary 50 (character string extraction Process 710). The character strings extracted in this manner correspond to the above-described related words, and these related words are respectively added to the related words dictionary 1 to 3 etc. corresponding to the external log 502 from which the related words are extracted. It is memorized.

  Various methods can be considered as methods for extracting related words from the external log 502. For example, the related words can be extracted by the method of the sentence analysis unit 43 described above.

  In the character string extraction process 710, related words are extracted from the plurality of external logs 502 and stored in the corresponding related word dictionary 50 respectively. For example, the plurality of external logs 502 may be text data collected at different timings for similar objects or data sources, or may be text data collected at different timings according to different subjects or search conditions. . The detailed process of the character string extraction process 710 will be described later.

  Next, the information updating unit 46 compares the plurality of related word dictionaries 50 in which the related words are stored by the character string extraction process 710, and compares the comparison result with the comparison result data 54 according to the appearance situation of the related words. It memorizes (dictionary comparison processing 720).

  For example, when comparing two related word dictionaries (related word dictionary 1 and related word dictionary 2) collected at different timings, if there is a related word which does not exist in related word dictionary 1 but exists in related word dictionary 2 This is stored in the comparison result data 54 as a newly appeared new relevance, and conversely, if there is a relevance that is present in the relevance dictionary 1 but not in the relevance dictionary 2, it is eliminated Are stored in the comparison result data 54 as related words.

  Also, for example, when comparing three related word dictionaries (related words 1 to 3) collected at the same timing for different subjects, if there are related words present in all related word dictionaries 1 to 3, , Are stored in the comparison result data 54 as common related words.

  In addition, it can be flexibly specified according to the use mode of the information updating unit 46 as to what appearance condition the related words are stored in the comparison result data 54 in the plurality of related word dictionaries. The detailed process of the dictionary comparison process 720 will be described in detail later.

  When the conversation control terminal device 2 ′ ′ requests the display of comparison result data 54 obtained by the user comparing the predetermined related words dictionary directly or according to the response scenario, the topic providing server 4 ′ performs these comparison results When input identification information including data 54 is transmitted to the conversation control terminal device 2 ′ ′ and the speech control terminal device 2 ′ ′ receives the input identification information, the response information is determined based on the input identification information and the scenario data 28. Control to display the response information on the display of the conversation control terminal device 2 ''.

  On the display of the conversation control terminal device 2 ′ ′, for example, the topic name and the change of the related word in this topic are displayed. The indication of the change in the related words includes, for example, the appearance status of the related words and the corresponding words corresponding thereto.

  With regard to comparison result data 54 in FIG. 46 described later, for example, regarding the topic name “topic of“ trade of stock ”at t10 on October 10, 2013”, the related word “tax rate” whose appearance status is “new arrival related word” And the related word "account" whose appearance status is "disappeared related word" are displayed. This display corresponds to the record 54a of the comparison result data 54 shown in FIG. The user sees the content displayed on the display, and a related word "tax rate" newly appears at the timing of t2 in the topic related to the topic name, and at the same time, the related word "account" disappears You can get noticed.

<< Description of character string extraction processing in the information update unit >>
The character string extraction process 710 will be described with reference to FIG. FIG. 40 is a flowchart showing the procedure of the character string extraction process 710. First, in step S61, an external log 502 (text data) to be extracted as related words is read. Text data may be any data as long as text data can be obtained, as described above.

  Next, in step S62, from the text data read in step S61, a related word which is a character string that can be semantically identified is extracted. There are various methods for extracting related words from text data, as described above, including a method based on the degree of difference between adjacent characters before and after by the sentence analysis unit 43, a method using morphological analysis, and the like.

  Next, in step S63, when a plurality of related words are extracted for one text data in step S62, the plurality of related words are ranked according to a predetermined determination criterion. For example, ranking can be performed according to the importance of related words in text data, and ranking can be performed according to the character length and the frequency of appearance of related words. Further, in the case of extracting related words by a method based on the degree of difference between adjacent characters before and after, ranking is performed according to the degree of difference between adjacent characters before and after. In addition, ranking can be performed by various elements other than such a standard, and a combination thereof. Such "rank" indicates the relevance to the topic. Also, even when a plurality of related words are extracted, such ranking can be avoided.

  Next, in step S64, the related words ranked in step S63 are stored in the related word dictionary corresponding to the text data. For example, related words extracted from one text data are collectively stored in one record, and each related word is stored in a storage location (sequence entry) according to the ranking. The related words are stored as described above, whereby a plurality of related words are defined as a set associated with one text data (text data from which those related words are extracted). Ranking is to rank related words in the set.

  In the character string extraction process, when there is a plurality of text data to be processed, the process from step S61 to step S64 described above is repeated for each text data.

<< Description of dictionary comparison processing in the information update unit >>
The dictionary comparison process 720 will be described with reference to FIG. FIG. 41 is a flowchart showing the processing procedure of the dictionary comparison processing 720. In this embodiment, related words are extracted from two text data (text data 1 and text data 2) collected as time series data, and the related word dictionary (i-1) and the related word dictionary (i) It is assumed that comparison processing is performed in the situation stored in.

  First, at step S71, the related words stored in the related word dictionary (i-1) and the related word dictionary (i) are read out. Next, in step S72, the related words of the related word dictionary (i-1) and the related word dictionary (i) are compared, and among them, the related words dictionary (i-1) does not exist, and the related word dictionary (i) Are stored in the comparison result data 54 as newly appearing related words (newly arrived related words). Each related-word dictionary is, for example, associated with a topic name, and the dictionary comparison processing 720 can perform comparison using this topic name. The newly arrived related word has the topic name that can represent the corresponding associated word dictionary, the appearance status (in this case, the “new arrival” character indicating that it has newly appeared, the code corresponding to this, etc.) and the comparison result data 54 Is stored in

  Next, in step S73, the related words of the related word dictionary (i-1) and the related word dictionary (i) are compared, and among them, the related words dictionary (i-1) exists and the related word dictionary (i) The relative words not existing in are stored in the comparison result data 54 as the deleted relative words (erasing related words). Each related word dictionary is, for example, associated with a topic name, and the disappearing related word corresponds to the topic name, the appearance status (in this case, the "disappeared" character indicating that it newly appeared, or this) Are stored in the comparison result data 54 together with the code etc.).

  Thereafter, in step S74, the content of the related word dictionary (i-1) is copied to the related word dictionary (i). This is because, at the next timing, the character string extraction process 710 prepares a related-word dictionary (i-1) for storing related words, and then, with this new related-word dictionary (i-1) , The related word dictionary (i) in which the content of the related word dictionary (i-1) is copied is compared by the dictionary comparison process 720.

  As described above, the character string extraction process 710 and the dictionary comparison process 720 are repeatedly executed at predetermined timings, but the detailed description will be described later. Further, the dictionary comparison processing 720 is repeatedly processed, so that the comparison result is stored in the comparison result data 54 at the processing timing, but when the comparison result is stored, it is stored before that. Whether the comparison results are to be deleted or stored cumulatively is determined according to the specification of the information search system 100 according to the present invention. Alternatively, the comparison result data 54 may be prepared separately for each of the dictionary comparison processes 720.

  Further, although omitted in this example, the related word dictionary (i-1) and the related word dictionary (i) can be compared, and common related words (common related words) can be stored in the comparison result data 54. In this case, for example, another related term (co-occurrence related term) stored together with the common related term in the related term dictionary (i-1) and the other related term stored together with the common related term in the related term dictionary (i) The information on the commonality can be stored in the comparison result data 54 by further comparing whether or not there is commonality with the related words of the (co-related related words).

  Furthermore, when comparing co-related related words as described above, information regarding commonality may be determined in consideration of the ranks associated with the co-related related words. For example, when co-related related words having high rank (highly important for the topic indicated by their related words) are common to the related word dictionary (i-1) and the related word dictionary (i), common related words Commonality can be appreciated more.

<<< Description of character string extraction processing and dictionary comparison processing in information update unit >>>
In FIG. 42, related words are extracted at different timings by the character string extraction process 710 from five text data (text data 1 to 5) collected in time series from the same WEB page, and the extracted related words are extracted. Are stored in corresponding relevancy dictionary (i-1) or relevancy dictionary (i) respectively, and then relevancy dictionary (i-1) and relevancy dictionary when relevancy dictionary (i) is updated The dictionary comparison process 720 is performed for (i), and an example in which these processes are performed periodically from time (T = t1) to (T = t5) is shown (time (T = time). I have omitted after t6).

  First, at time (T = t1), from the text data 1 collected from the predetermined WEB page at this point in time, the related words are extracted by the character string extraction process 710a, and the extracted related words are the related words dictionary (i -1) is stored. The character string extraction processing 710a corresponds to the character string extraction processing 710 described with reference to FIG.

  At the next time (T = t2), as in the case of T = t1, from the text data 2 collected from the same web page, the related words are extracted by the character string extraction process 710b, and the extracted related words are It is stored in the related word dictionary (i). Here, if there is a change in the topic or the described content in the target WEB page, the associated words to be extracted will also change accordingly. When the related words are stored in the related words dictionary (i) (when updated), the dictionary comparison process 720a compares the related words dictionary (i-1) with the related words dictionary (i). In accordance with the appearance status, for example, a newly arrived related term etc. that has newly appeared is stored in the comparison result data 54. Also, when the comparison process is completed, the content of the related word dictionary (i) is copied to the related word dictionary (i-1).

  The dictionary comparison process 720a corresponds to the dictionary comparison process 720 described with reference to FIG. Although the dictionary comparison process 720a is described as being performed at the timing of T = t2 in this figure, it is performed after the related word dictionary (i) is updated.

  At the next time (T = t3), as in the case of T = t1 and t2, from the text data 3 collected from the same web page, the related words are extracted by the character string extraction process 710c and extracted The lyrics are stored in the related vocabulary dictionary (i). Here, if there is a change in the topic or the described content in the target WEB page, the associated words to be extracted will also change accordingly. When the related words are stored in the related words dictionary (i) (when updated), the dictionary comparison processing 720b compares the related words dictionary (i-1) with the related words dictionary (i). In accordance with the appearance status, for example, a newly arrived related term etc. that has newly appeared is stored in the comparison result data 54. Also, when the comparison processing is completed, the content of the related word dictionary (i) is copied (saved) to the related word dictionary (i-1).

  Thereafter, the character string extraction process (710d, 710e) and the dictionary comparison process (720c, 720d) are similarly repeated, and the comparison result data 54 includes the related word dictionary (i-1) and the related word dictionary (i). The relation words obtained as a result of comparison are updated at each timing (T = t2 to t5). As described above, whether the relative words as the comparison result are stored only in the latest one or in a cumulative manner depends on the specification of the applied system or the like to be applied.

  In FIG. 43, related words are extracted at different timings by the character string extraction process 710 from five text data (text data 1 to 5) collected in chronological order from the same WEB page, and the extracted related words are extracted. , The related word dictionary (i-1), the related word dictionary (i), or the related word dictionary (i + 1), and then the related word dictionary (i + 1) is updated, A dictionary comparison process 720 is performed on i-1), the related word dictionary (i) and the related word dictionary (i + 1), and these processes are cyclic from time (T = t1) to (T = t5) An example is shown in (The time (T = t6) and later is omitted). The difference from FIG. 41 is that three related word dictionaries are used cyclically.

  First, at time (T = t1), from the text data 1 collected from the predetermined WEB page at this point in time, the related words are extracted by the character string extraction process 710f, and the extracted related words are extracted from the related words dictionary (i -1) is stored.

  At the next time (T = t2), as in the case of T = t1, from the text data 2 collected from the same web page, the related words are extracted by the character string extraction process 710g, and the extracted related words are It is stored in the related word dictionary (i). Here, if there is a change in the topic or the described content in the target WEB page, the associated words to be extracted will also change accordingly. Thereafter, at time (T = t3), as in the case of T = t1 and t2, from the text data 3 collected from the same WEB page, the related words are extracted by the character string extraction process 710h, and the extracted related words Are stored in the related word dictionary (i + 1). Here, if there is a change in the topic or the described content in the target WEB page, the associated words to be extracted will also change accordingly.

  When related words are stored in the related word dictionary (i + 1) (when updated) at time (T = t3), three related word dictionaries (related word dictionary (i-1) The lyric dictionary (i) and the related lexicon dictionary (i + 1) are compared, and the related words are stored in the comparison result data 54 according to the appearance status of the related words. In this embodiment, since three related dictionaries are compared, as related words stored based on changes related to two related words dictionaries, for example, new newly appeared related words and annihilated delete related words, Some relative words are determined to be stored based on changes across the three relative word dictionaries. For example, it is a resurrected related word which has once disappeared in the related word dictionary (i) and reappeared in the related word dictionary (i + 1). In addition, it is a condition that the frequency of appearance of the related words is stored in the related words dictionary, but the related words whose appearance frequency is rapidly increased (in the three related words dictionary) in a short period of time (3 relations Maintain a certain range of appearance frequency as the appearance frequency of related words whose appearance frequency has been reduced (in the lexical dictionary), related words whose appearance frequency has become high again (in the three related word dictionaries), and other related words change It is also possible to grasp related words and the like.

  In the dictionary comparison process 720f, when the comparison process is completed, the contents of the related word dictionary (i) are copied to the related word dictionary (i-1), and the contents of the related word dictionary (i + 1) Copy to i). In this figure, although the dictionary comparison process 720f is described as being performed at the timing of T = t3, the related word dictionary (i-1) and the related word dictionary (i) are updated. It will be done later.

  At the next time (T = t4), as in the case of T = t1 to t3, the related words are extracted by the character string extraction process 710i from the text data 4 collected from the same WEB page and extracted The lyrics are stored in the related vocabulary dictionary (i + 1). Here, if there is a change in the topic or the described content in the target WEB page, the associated words to be extracted will also change accordingly. When related words are stored in the related word dictionary (i + 1) (when updated), three related word dictionaries (related word dictionary (i-1), related word dictionary (i), related words by the dictionary comparison process 720g) The dictionary (i + 1) is compared, and the related words and the like are stored in the comparison result data 54 according to the appearance status of the related words. Also, when the comparison process is completed, the contents of the related word dictionary (i) are copied to the related word dictionary (i-1), and the contents of the related word dictionary (i + 1) are copied to the related word dictionary (i) Be done.

  Thereafter, the character string extraction process 710j and the dictionary comparison process 720h are similarly repeated, and the comparison result data 54 compares the related word dictionary (i-1), the related word dictionary (i), and the related word dictionary (i + 1). It is updated at each timing (T = t3 to t5) by the related words obtained as a result. As described above, whether the relative words as the comparison result are stored only in the latest one or in a cumulative manner depends on the specification of the applied system or the like to be applied.

  In the embodiment of FIG. 42, the comparison result data 54 is stored and updated using (two cyclically) two relevancy dictionaries and three relevancy dictionaries (cyclically) in the embodiment of FIG. Dictionary comparison processing may be performed using a related-word dictionary. This makes it possible to grasp the appearance status of related words at more timings, and when the appearance status satisfies a predetermined condition, the related terms can be stored in the comparison result data 54.

  In FIG. 44, related words are extracted by the character string extraction process 710 from three text data (text data A to C) collected at the same timing from different web pages (web pages for different subjects). The lyrics are stored in corresponding relevancy dictionary A, relevancy dictionary B, or relevancy dictionary C, and then the dictionary comparison process 720 is performed on these three relevancy dictionaries, and these processes are performed as follows: The example shown periodically from time (T = t1) to (T = t3) is shown (after time (T = t4) is omitted).

  First, at time (T = t1), character string extraction processing 710k, character string extraction processing 710m, characters from three text data (text data A to C) collected respectively from predetermined different WEB pages at this point in time The related words are extracted by the column extraction process 710 n, and the extracted related words are stored in the related word dictionary A, the related word dictionary B, and the related word dictionary C, respectively.

  Thereafter, comparison of three related words dictionaries (related words dictionary A, related words dictionary B, related words dictionary C) is performed by the dictionary comparison processing 720k, and the related words etc. are compared result data according to the appearance situation of the related words It is stored in 54. In this embodiment, for example, related words common to three related word dictionaries are stored in the comparison result data 54. In this case, the text data A to C are text data collected from the web pages on different subjects, and it is better to focus on the related words common to the three related words dictionaries (common related words) rather than different related words. It is possible to discover common topics, and in many cases it makes sense.

  Also, other related words (co-occurrence related words) stored with common related words in related word dictionary A, and other related words stored with common related words in related word dictionary B (co-related related words), And whether or not there is commonality with other related words (co-relevant related words) stored together with the common related word in the related word dictionary C and information related to the commonness as a comparison result data 54 can be stored.

  Furthermore, when comparing co-related related words as described above, information regarding commonality may be determined in consideration of the ranks associated with the co-related related words. For example, when co-related related words having high rank (highly important for the topic indicated by their related words) are common in the three related word dictionaries, the commonality of common related words can be more highly evaluated.

  Such comparison of three related words dictionaries at time (T = t1) can be repeated at time (T = t2). By performing such processing, the comparison result data 54 can be updated in time series.

  At time (T = t2), similar to time (T = t1), character string extraction processing from three text data (text data A ′ to C ′) collected respectively from predetermined different WEB pages at this time Related words are extracted by 710k ′, character string extraction process 710m ′, and character string extraction process 710n ′, and the extracted related words are stored in the related word dictionary A ′, the related word dictionary B ′, and the related word dictionary C ′, respectively. Be done. In this embodiment, the text data A ′ is assumed to be the same web page as the text data A, or a web page of the same subject. Similarly, the text data B 'is the same WEB page as the text data B or the WEB page of the same subject, and the text data C' is the WEB page the same as the text data C or the WEB page of the same subject.

  Thereafter, the dictionary comparison processing 720k 'compares three related words dictionaries (related words dictionary A', related words dictionary B ', related words dictionary C'), and according to the appearance situation of the related words, the related words etc. Are stored in the comparison result data 54. In this embodiment, for example, related words common to three related word dictionaries are stored in the comparison result data 54.

  Furthermore, at time (T = t3), as with time (T = t1, t2), three pieces of text data (text data A ′ ′-C ′ ′) collected respectively from predetermined different WEB pages at this time Related words are extracted from the character string extraction process 710k ′ ′, the character string extraction process 710m ′ ′, and the character string extraction process 710n ′ ′, and the extracted related words are the related word dictionary A ′ ′, the related word dictionary B ", stored in the related vocabulary dictionary C". In this embodiment, the text data A ′ ′ is assumed to be the text data A, the same web page as the text data A ′, or the web page of the same subject. Similarly, the text data B ′ ′ is the same web page as the text data B, the text data B ′, or the web page of the same subject, and the text data C ′ ′ is the same web as the text data C, the text data C ′. It is a page or a web page of the same subject.

  Thereafter, the dictionary comparison process 720k ′ ′ compares three related word dictionaries (related word dictionary A ′ ′, related word dictionary B ′ ′, related word dictionary C ′ ′) according to the appearance situation of the related words Related words and the like are stored in the comparison result data 54. In this embodiment, for example, related words common to three related word dictionaries are stored in the comparison result data 54.

  In the example of FIG. 44, related words are extracted based on three text data collected from different WEB pages (WEB pages related to different subjects) at the same timing. The words may be extracted, or the related words may be extracted from four or more pieces of text data.

<< Detailed Description of Character String Extraction Processing and Dictionary Comparison Processing in Information Update Unit >>
Next, an example of the character string extraction process 710 and the dictionary comparison process 720 shown in FIG. 43 will be described in more detail with reference to FIG. In FIG. 45, character string extraction processing (710f, 710g, 710h) is performed on three pieces of text data (text data 1 to 3), and the corresponding lexicon dictionary (i-1) and the related lexicon dictionary (i) ), The dictionary comparison process 720 f is performed on the related word dictionary (i + 1).

  In this example, the three text data are collected from the same web page associated with the common subject "stock trading". For example, the search keyword "trade of stock" is input by web search, and the three web pages obtained as a result are treated as one text data. In FIG. 45, the text data 1 is the text data 1-1 obtained from the first WEB page, the text data 1-2 obtained from the second WEB page, and the third at time (T = t1). It includes text data 1-3 obtained from the WEB page. Similarly, text data 2 includes text data 2-1 obtained from the first WEB page, text data 2-2 obtained from the second WEB page, and a third WEB at time (T = t2). The text data 2-3 obtained from the page, and the text data 3 is obtained from the text data 3-1 obtained from the first WEB page and the second WEB page at time (T = t3) Text data 3-2 and text data 3-3 obtained from the third WEB page are included. Here, the URLs of the first WEB page are all the same, the URLs of the second WEB page are all the same, and the URLs of the third WEB page are all the same.

  Here, text data (text data 1-1, text data 1-2, text data 1-3) corresponding to three web pages included in one text data are respectively converted into question sentences as shown in FIG. 26. It can be considered in association. For example, the text data 1-1 is a question of Q 1, the text data 1-2 is a question of Q 8, and the text data 1-3 is a question of Q 13.

  The character string extraction process 710 f extracts related words from the text data 1 according to a predetermined method at time (T = t 1), and stores the related words in the related word dictionary (i−1). In this embodiment, four related words are extracted, each of which is arranged in order of rank and stored as one record in the related word dictionary (i-1). The ranking of the related words can be determined based on, for example, the frequency of appearance. The extracted four related words (related words 1 to 4) are “○ company”, “Δ bank”, “application”, and “account” in order of rank. In morpheme analysis, etc., the character string is grasped by breaking it down into the smallest unit (morpheme) having meaning, but in other methods, a unit larger than the morpheme (eg, a sentence or a part of a sentence) is extracted as a related word sell. Character strings consisting of nouns and particles, such as the above-mentioned "application", are also extracted as related words.

  Further, in this embodiment, four related words are extracted for each text data, but this is for the convenience of explanation (the same applies to the following embodiments). In practice, any number of related words may be extracted, and the number varies depending on text data. In the character string extraction process, when the character string is determined as a related word when the threshold value is exceeded in a certain judgment standard, the number of related words extracted changes according to the judgment. Further, the number of related words to be extracted can be fixed, or can be set according to the number of characters of text data.

  The character string extraction process 710g extracts related words from the text data 2 according to a predetermined method at time (T = t2), and stores the related words in the related word dictionary (i). In this embodiment, four related words are extracted, each of which is arranged in order of rank and stored as one record in the related word dictionary (i). The ranking of the related words can be determined based on, for example, the frequency of appearance. The extracted four related words (related words 1 to 4) are “tax rate”, “company ○”, “bank”, and “application” in order of rank.

  Similarly, at time (T = t3), the character string extraction process 710h extracts related words from the text data 3 according to a predetermined method, and stores the related words in the related word dictionary (i + 1). In this embodiment, four related words are extracted, each of which is arranged in the order of rank and stored as one record in the related word dictionary (i + 1). The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) extracted are “○ company”, “account”, “Δ bank”, and “application” in the order of rank.

  Next, dictionary comparison processing 720 f is performed on the related word dictionary (i−1), the related word dictionary (i), and the related word dictionary (i + 1). In this embodiment, the dictionary comparison process 720f detects newly appeared related words (newly arrived related words), eliminated related terms (disappeared related terms), and re-appeared related terms (resurrection related terms), and detects them. It is assumed that the comparison result data 54 is stored.

  For example, when the related word dictionary (i-1) and the related word dictionary (i) are compared, in the related word dictionary (i), the related word "tax rate" newly appears, and further, the related word "account" is It has disappeared. Therefore, as shown in the record 54a of FIG. 46, the related words "tax rate" and "account" are stored in the comparison result data 54. In addition, in the comparison result data 54, the data representing the appearance status (in this embodiment, "new arrival" in the case of a newly appearing related word, "annihilation" in the case of a deleted related word) is the same. It is stored in the record. Furthermore, in this embodiment, in order to indicate the timing at which the appearance status has been reached, data of "topic name" for identifying the related word dictionary is stored. Each associated word dictionary is associated with a topic name and date, and the associated word dictionary (i) is associated with a topic name such as "the topic of" trade of stock "at t10, October 10, 2013" here. It is done.

  Next, when the related word dictionary (i) and the related word dictionary (i + 1) are compared, the related word "tax rate" disappears in the related word dictionary (i + 1), and the related word "account" is restored. (It exists in the related word dictionary (i-1) and has disappeared in the related word dictionary (i)). Therefore, as shown in the record 54b of FIG. 46, the related words "account" and "tax rate" are stored in the comparison result data 54. In addition, in the comparison result data 54, data indicating the appearance status together with these related words (in this example, "revived" in the case of the related words that reappeared (revived), and "extinguished" in the case of the deleted related words ) Are stored in the same record. Furthermore, in this embodiment, in order to indicate the timing at which the appearance status has been reached, data of "topic name" for identifying the related word dictionary is stored. Each related-word dictionary is associated with a topic name and date, and the related-word dictionary (i + 1) is associated with a topic name such as "the topic of" share trading "at t10, October 10, 2013" here. It is done.

  Next, the example of the character string extraction process 710 and the dictionary comparison process 720 shown in FIG. 44 will be described in more detail with reference to FIG. In FIG. 47, character string extraction processing (710k, 710m, 710n) is performed on three text data (text data A to C) at time (T = t1), and the corresponding relevancy dictionary A, association is performed. It shows that dictionary comparison processing 720k is performed for the lyric dictionary B and the related lexicon dictionary C.

  In this example, the three text data are collected from the web pages associated with different subjects at the same time (T = t1). That is, text data A is based on a web page having "the technology of company A" as the subject, and text data B is based on a web page having "the technology of company B" as the subject, and text data C is based on the web page which made "AI (artificial intelligence) related technology" the subject.

  For example, with regard to text data A, a search keyword “technology of company A” is input by web search, and the three web pages obtained as a result are treated as one text data. Similarly, for the text data B, the search keyword “technology of company B” is input by web search, the three web pages obtained as a result are treated as one text data, and for the text data C, the web search The search keyword “AI (artificial intelligence) related technology” is input, and the three web pages obtained as a result are treated as one text data.

  In FIG. 47, the text data 1 is text data A-1 obtained from the first WEB page and text data A-2 obtained from the second WEB page, which are related to the subject "technology of company A". It includes text data A-3 obtained from the third web page. Similarly, text data 2 includes text data B-1 obtained from the first WEB page, text data B-2 obtained from the second WEB page, and the like related to the subject "the technology of company B". 3 includes the text data B-3 obtained from the third web page, and the text data 3 is the text data C-1 obtained from the first web page related to the subject "AI (artificial intelligence) related technology", It includes text data C-2 obtained from the second WEB page and text data C-3 obtained from the third WEB page.

  The character string extraction process 710k extracts related words from the text data A according to a predetermined method, and stores the related words in the related word dictionary A. In this embodiment, four related words are extracted, each of which is arranged in order of rank, and is stored in the related word dictionary A as one record. The ranking of the related words can be determined based on, for example, the frequency of appearance. The extracted four related words (related words 1 to 4) are "A company", "speech", "speech recognition" and "sales" in order of rank.

  The character string extraction process 710 m extracts related words from the text data B according to a predetermined method, and stores the related words in the related word dictionary B. In this embodiment, four related words are extracted, each of which is arranged in order of rank, and is stored in the related word dictionary B as one record. The ranking of the related words can be determined based on, for example, the frequency of appearance. The extracted four related words (related words 1 to 4) are “voice”, “research and development”, “performance by company B”, and “voice recognition” in order of rank.

  Similarly, the character string extraction process 710 n extracts related words from the text data C according to a predetermined method, and stores the related words in the related word dictionary C. In this embodiment, four related words are extracted, each of which is arranged in order of rank, and is stored in the related word dictionary C as one record. The ranking of the related words can be determined based on, for example, the frequency of appearance. The extracted four related words (related words 1 to 4) are “AI”, “robot”, “speech recognition”, and “agent” in order of rank.

  Next, dictionary comparison processing 720k is performed on the related word dictionary A, the related word dictionary B, and the related word dictionary C. In this embodiment, the dictionary comparison process 720k detects related words (common related words) common to the three dictionaries and stores them in the comparison result data 54.

  When comparing the related words dictionary A, the related words dictionary B, and the related words dictionary C, the related words described above are stored respectively, and it is recognized that "speech recognition" exists as a common related word. Therefore, this "voice recognition" is stored in the comparison result data 54.

  By grasping such common related words, analysis of inter-company related information can be realized effectively. For example, the related words often used in the description of the technology of company A are extracted by the character string extraction process 710k based on text data A, and the related words often used in the description of technology of company B are The company A and company B are extracted by the character string extraction process 710 n based on the text data C, which is extracted by the character string extraction process 710 m based on and is used frequently in the description related to the AI (artificial intelligence) related technology. However, it is possible to objectively grasp what kind of AI related technology may have commonality.

  In the example shown in FIG. 47, relative word dictionaries based on text data collected from web pages related to different subjects are compared at the same time (T = t1), and each web page is Although it is collected by intentional WEB search, as a result of comparing the related words dictionary obtained respectively from the WEB page collected completely by chance, it may be considered that a common related word is found also by chance .

  Next, another embodiment of the character string extraction process 710 and the dictionary comparison process 720 in the information updating unit 46 will be described with reference to FIG. FIG. 48 shows a modification of the character string extraction process 710 and the dictionary comparison process 720 shown in FIG. In FIG. 48, character string extraction processing 710 is performed on each of two text data (text data 1 and text data 2), and the corresponding lexicon dictionary (i-1) and relating dictionary lexicon (i) are obtained. , A dictionary comparison process 720 is performed. The display of the process related to text data 3 shown in FIG. 45 is omitted.

  In this example, the two text data are collected from the same web page associated with the common subject "stock trading". For example, the search keyword “trade dealings” is input by web search, and the three web pages obtained as a result are regarded as one text data, but the related words are extracted for each web page unit. As a result, although related words are managed for each WEB page, three text data based on three Web pages may be prepared, and related words may be extracted for each text data. Here, it is important that related words are extracted from a plurality of text data.

  In FIG. 48, the text data 1 is the text data 1-1 obtained from the first WEB page, the text data 1-2 obtained from the second WEB page, and the third at time (T = t1). It includes text data 1-3 obtained from the WEB page. Similarly, text data 2 includes text data 2-1 obtained from the first WEB page, text data 2-2 obtained from the second WEB page, and a third WEB at time (T = t2). It contains text data 2-3 obtained from the page. Here, the URLs of the first WEB page are all the same, the URLs of the second WEB page are all the same, and the URLs of the third WEB page are all the same.

  The character string extraction process 710 extracts related words from the text data 1 according to a predetermined method at time (T = t1). As described above, this is performed for each text data included in the text data 1. For example, related words are extracted from the text data 1-1 by a predetermined method, and related words are extracted from the text data 1-2 by a predetermined method. The words are extracted, and the related words are extracted from the text data 1-3 by a predetermined method. In this embodiment, four related words are extracted for each, and each related word is arranged in order of rank. The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) related to the text data 1-1 are, in order of rank, "○ company", "account", "application procedure", and "Δ bank". The four related words (related words 1 to 4) related to the text data 1-2 are "account", "bank", "application", and "company" in order of rank. The four related words (related words 1 to 4) related to the text data 1-3 are "purchase of stock", "bank", "limit price", and "company ○" in order of rank.

  Next, the character string extraction process 710 obtains a neighborhood relevancy for each of the relevancy thus extracted, and stores them in the relevancy dictionary (i-1). The neighborhood related words are related words that appear (co-occur) with the related words when focusing on a certain related word. In each text data (1-1, 1-2, 1-3), a set of related words included in a topic corresponding to the text data is a related word set, and when focusing on a certain related word, A set of related words including related words is referred to as a neighborhood of related words, and a set of neighborhoods of related words is referred to as a neighborhood system of related words. The related-word dictionary (i-1) stores, for each related word, a neighborhood system of related words.

  For example, when focusing on the related word "○ company", this related word is extracted for the text data 1-1, and the neighborhood of the related word is {は company, account, application procedure, 銀行 bank}. Similarly, the related words are extracted for the text data 1-3, and the neighborhoods of the related words are {share purchase, Δ bank, limit price, ○ company}. From these, with regard to the related word “company ○,” the neighborhood system of the related word is {company ○, account, purchase of stock, bank, application procedure, limit price} (near the related word for text data 1-1 And the related words “△ bank” overlapping in the vicinity of the related words for the text data 1-3 are included only one).

  The neighborhood systems of the related words thus obtained are the respective related words "company Y", "account", "bank", "application procedure", "purchase of stock", "application", "limit price", " "Company" is stored in the related word dictionary (i-1). For each related word, a neighborhood system of related words (near-point related words 1 to 7) is stored, but the order of these related words is the ranking performed by the character string extraction process 710, the height of co-occurrence, etc. It is determined in consideration of

  Furthermore, the character string extraction process 710 similarly extracts related words from the text data 2 according to a predetermined method at time (T = t2). As described above, this is performed for each text data included in the text data 2. For example, related words are extracted from the text data 2-1 by a predetermined method, and related words are extracted from the text data 2-2 by a predetermined method. The words are extracted, and the related words are extracted from the text data 2-3 by a predetermined method. In this embodiment, four related words are extracted for each, and each related word is arranged in order of rank. The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) related to the text data 2-1 are, in order of rank, "company ○", "account", "new system", and "application procedure". The four related words (related words 1 to 4) related to the text data 2-2 are “account”, “Δ bank”, “◇ company”, and “purchase of stock” in order of rank. Regarding the text data 2-3, the four related words (related words 1 to 4) extracted are “purchase of stock”, “bank”, “company O” and “new system” in order of rank.

  Next, the character string extraction process 710 obtains a neighborhood relevancy for each of the relevancy thus extracted, and stores them in the relevancy dictionary (i). For example, focusing on the related word "○ company", this related word is extracted for the text data 2-1, and the neighborhood of the related word is {○ company, account, new system, application procedure}. Similarly, the related words are extracted for the text data 2-3, and the neighborhoods of the related words are {share purchase, Δ bank, company ○, new system}. From these, with regard to the related word "○ company", the neighborhood system of the related word becomes {○ company, account, stock purchase, new system, application procedure, △ bank} (related word of text data 2-1) Only one relevant term “new system” is included in the neighborhood and the neighborhood of the relative term for the text data 2-3).

  The neighborhood system of related words found in this way is the respective related words "○ company", "account", "△ bank", "application procedure", "share purchase", "new system", "◇ company" , Related words dictionary (i) is stored. For each related word, a neighborhood system of related words (near-point related words 1 to 6) is stored, but the order of these is determined by the ranking performed by the character string extraction process 710, the height of co-occurrence, etc. It is determined in consideration of

  Next, the dictionary comparison process 720 compares the related word dictionary (i-1) with the related word dictionary (i). As a result, the related words "limit price" and "apply" are the extinct related words that disappear at time (T = t2) (see reference numeral 753 in FIG. 48), and the related word "new system" is time (T =). The new arrival related words that newly appear at t2) (see reference numeral 752 in FIG. 48), these related words are compared as the change in the related word dictionary (i-1) and the related word dictionary (i). Is stored in

  Furthermore, the dictionary comparison process 720 is performed for the relevant words “company 、”, “account”, “△ bank”, “application procedure”, “purchase of stock”, and “◇ company”, with time (T = t1) and time Although it is related words which exist in both of (T = t2), comparison is also performed with respect to neighborhood related words of each related word. In this case, it can be understood that the neighborhood related words (or the order of the neighborhood related words) are changed (see reference numeral 751 in FIG. 48). This indicates that the relationship and the position of the relevant words in the topic are changing. In the case where the information search system 100 according to the present invention tries to visualize such changes in the proximity related words, such information is also stored in the comparison result data 54.

  Next, another embodiment of the character string extraction process 710 and the dictionary comparison process 720 in the information updating unit 46 will be described with reference to FIG. FIG. 49 shows a modification of the character string extraction process 710 and the dictionary comparison process 720 shown in FIG. In FIG. 49, character string extraction processing 710 is performed on each of three text data (text data A to C), and the corresponding relevancy dictionary A, relevancy dictionary B, and relevancy dictionary C A comparison process 720 is shown to be performed.

  In this example, the three text data are collected at the same time (T = t1) from the same WEB page associated with different subjects. That is, text data A is based on a web page having "the technology of company A" as the subject, and text data B is based on a web page having "the technology of company B" as the subject, and text data C is based on the web page which made "AI (artificial intelligence) related technology" the subject.

  For example, with regard to text data A, search keyword "technology of company A" is input by web search, and two text data (text data A-1, text data A-2) are obtained from the two web pages obtained as a result. ) And handle them individually in the string extraction process 710. Similarly, for the text data B, the search keyword “technology of company B” is input by web search, and two text data (text data B−1, text data B−) are obtained from the two web pages obtained as a result. 2) are obtained, and these are handled individually in the character string extraction process 710. Similarly, for the text data C, the search keyword "AI (artificial intelligence) related technology" is input by web search, and two text data (text data C-1, text) are obtained from the two web pages obtained as a result. Data C-2) is acquired, and these are handled individually in the character string extraction process 710. In FIG. 48, each of the text data A, the text data B, and the text data C includes three text data, but in this embodiment, it is assumed that each includes two text data.

  Character string extraction processing 710 extracts related words from text data A according to a predetermined method at time (T = t1). As described above, this is performed for each text data included in text data A. For example, related words are extracted from text data A-1 by a predetermined method, and related words are extracted from text data A-2 by a predetermined method. The lyrics are extracted. In this embodiment, four related words are extracted for each, and each related word is arranged in order of rank. The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) related to the text data A-1 are "company A", "speech", "speech recognition" and "robot" in order of rank. The four related words (related words 1 to 4) related to the text data A-2 are "compression technology", "speech recognition", "sales", and "speech" in order of rank.

  Next, the character string extraction processing 710 obtains a neighborhood relevancy for each of the relevancy thus extracted, and stores them in the relevancy dictionary A. The neighborhood related words are related words that appear (co-occur) with the related words when focusing on a certain related word. In each text data (A-1, A-2), a set of related words included in a topic corresponding to the text data is a related word set, and when focusing on a certain related word, the related words are included. A set of related words is referred to as a neighborhood of related words, and a set of neighborhoods of related words is referred to as a neighborhood system of related words. The related-word dictionary A stores a neighborhood system of related words for each related word.

  For example, focusing on the related word "speech recognition", this related word is extracted for the text data A-1, and the neighborhood of the related word is {A company, speech, speech recognition, robot}. Similarly, the related words are extracted for the text data A-2, and the neighborhoods of the related words are {compression technology, speech recognition, sales, speech}. From these, with regard to the related word "speech recognition", the neighborhood system of the related word becomes {speech recognition, company A, compression technology, speech, robot, sales} (near the related word for text data A-1, Only one relevant term "speech" overlapping with the neighborhood of the relevant term for text data A-2 is included).

  The neighborhood system of the related words thus obtained is stored in the related word dictionary A for each of the related words "A company", "speech recognition", "speech", "compression technology", "sales" and "robot". Ru. For each related word, the neighborhood system of the related word (near-point related words 1 to 5) is stored, but the order of these is determined by the ranking performed by the character string extraction process 710, the height of co-occurrence, etc. It is determined in consideration of

  Furthermore, the character string extraction process 710 extracts related words from the text data B according to a predetermined method at time (T = t1). As described above, this is performed for each text data included in text data B. For example, related words are extracted from text data B-1 by a predetermined method, and related words are extracted from text data B-2 by a predetermined method. The lyrics are extracted. In this embodiment, four related words are extracted for each, and each related word is arranged in order of rank. The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) related to the text data B-1 are "voice", "performance of company B", "speech recognition", and "research and development" in order of rank. The four related words (related words 1 to 4) related to the text data B-2 are "research and development", "speech", "speech recognition" and "authentication technology" in order of rank.

  Next, the character string extraction process 710 obtains a neighborhood relevancy for each of the relevancy thus extracted, and stores them in the relevancy dictionary B. For example, focusing on the related word "speech", this related word is extracted for the text data B-1, and the neighborhood of the related word is {speech, performance of company B, speech recognition, research and development}. Similarly, the related words are extracted for the text data B-2, and the vicinity of the related words is {research and development, speech, speech recognition, authentication technology}. From these, with regard to the related word "voice", the neighborhood system of the related word becomes {voice, research and development, performance of company B, speech recognition, authentication technology} (near the related word for text data B-1, The related words “speech recognition” and “research and development” overlapping with the neighborhood of the related words in the text data B-2 may be included only one each).

  The neighborhood system of the related words thus obtained is stored in the related word dictionary B for each of the related words "speech", "research and development", "performance by company B", "speech recognition" and "authentication technology". . For each related word, a neighborhood system of related words (near-point related words 1 to 4) is stored, but the order of these is determined by the ranking performed by the character string extraction process 710, the height of co-occurrence, etc. It is determined in consideration of

  Furthermore, the character string extraction processing 710 extracts related words from the text data C by a predetermined method at time (T = t1). As described above, this is performed for each text data included in the text data C. For example, related words are extracted from the text data C-1 by a predetermined method, and related words are extracted from the text data C-2 by a predetermined method. The lyrics are extracted. In this embodiment, four related words are extracted for each, and each related word is arranged in order of rank. The ranking of the related words can be determined based on, for example, the frequency of appearance. The four related words (related words 1 to 4) related to the text data C-1 are “AI”, “agent”, “robot”, and “voice recognition” in order of rank. The four related words (related words 1 to 4) related to the text data C-2 are “robot”, “speech recognition”, “AI”, and “learning function” in order of rank.

  Next, the character string extraction processing 710 obtains a neighborhood relevancy for each of the relevancy thus extracted, and stores them in the relevancy dictionary C. For example, focusing on the related word "AI", this related word is extracted for the text data C-1, and the neighborhood of the related word is {AI, agent, robot, speech recognition}. Similarly, the related words are extracted for the text data C-2, and the vicinity of the related words is {robot, speech recognition, AI, learning function}. From these, with regard to the related word “AI”, the neighborhood system of the related word becomes {AI, robot, agent, speech recognition, learning function} (near the related word for the text data C−1 and the text data C− The relevant words “robot” and “speech recognition” overlapping with the neighborhood of the related word of 2 are included only one each).

  The neighborhood system of the related words thus obtained is stored in the related word dictionary C for the respective related words “AI”, “robot”, “speech recognition”, “agent”, and “learning function”. For each related word, a neighborhood system of related words (near-point related words 1 to 4) is stored, but the order of these is determined by the ranking performed by the character string extraction process 710, the height of co-occurrence, etc. It is determined in consideration of

  Next, the dictionary comparison process 720 compares related word dictionaries A to C. As a result, the related word "speech recognition" is a related word (common related word) common to the three related word dictionaries at time (T = t1) (see the symbols 755, 755 'and 755' 'in FIG. 49). And this is stored in the comparison result data 54. By grasping such common related words, it is possible to discover common topics from different themes, and setting the subjects as in this embodiment also leads to analysis of inter-company related information.

  Also, the dictionary comparison processing 720 compares the common relatives with the neighborhood relatives of the respective relatives. In this way, it is possible to grasp the commonality of the near related words, the commonness of the order of the near related words, and the like, and it is possible to determine the level of commonness between the common related words.

  Although "speech" can be grasped as a common related word in the related word dictionary A and the related word dictionary B (see the symbols 756 and 756 'in FIG. 49), the common related words in such a part of the related word dictionary are also As it may be important information, it can be stored in the comparison result data 54.

<<< Context learning function and service ID switching >>>
In the information search system 100 according to an embodiment of the present invention, in the dialogue between the conversation control terminal device 2 '' and the user who uses this device, as described above, the input index, the progress index, etc. are learned. Dialogue can be realized according to the input index or the progress index. Such a function is a context learning function. The input index is information indicating what kind of input the user has made so far, that is, the history of the user's input. The progress indicator is information indicating what kind of topic has been provided to the user so far, that is, the history of the topic provided to the user.

  Furthermore, in the information retrieval system 100, by capturing the history or the like regarding the appearance of the related words in the information updating unit 46 of the topic providing server 4 ′, the topic name to which the related words belong can be clarified. It is possible to distinguish and handle lyrics and (newly appeared) new arrival related words, or to perform judgment on similarities and differences of topics by comparison processing of related words dictionary, and so on Function can be realized. With such a related word learning function, the user of the conversation control terminal device 2 '' can handle input types related to many topic names, and the input identification means of the user can be diversified.

  The scenario data is controlled to switch the service ID based on the input type determined from the input situation in the user's conversation control terminal device 2 ′ ′. For example, the related word dictionary 50 and related words / co-occurring words list display screen 650 provided by the sentence analysis unit 43, the preference analysis unit 44, and the topic analysis unit 45 of the topic providing server 4 'in the information search system 100 are provided. Service ID corresponding to the function to be selected and the service ID corresponding to the function providing the display of the related term dictionary 50 and the comparison result data 54 provided by the information updating unit 46 of the topic providing server 4 Switch.

The statement of the corresponding scenario data is set to transition to the service of the corresponding service ID, for example, as an action when a predetermined input type is input. If this is expressed by scenario data as shown in FIG. 14, the following statement is obtained.
<sto: $ IDN $: <sta: $ num $>: $ input $>
Here, “sto” is a description (shift to) for causing a state transition, “$ IDN $” is a service identification number of the transition destination, and “<sta: $ num $>” is the transition destination. Is a state number in the service of “<$ input $>” is a user input statement.

<<< Description of Hardware Configuration of Topic Providing Server According to One Embodiment of the Present Invention >>>
Next, with reference to FIG. 50, an example of a hardware configuration of a computer configuring the topic providing server 4 'according to an embodiment of the present invention will be described. However, the configuration of the topic providing server 4 'shown in FIG. 50 merely exemplifies the typical configuration.

  The topic providing server 4 ′ includes a central processing unit (CPU) 801, a random access memory (RAM) 802, a read only memory (ROM) 803, a network interface 804, an audio control unit 805, a microphone 806, a speaker 807, a display controller 808, A display 809, an input device interface 810, a keyboard 811, a mouse 812, an external storage device 813, an external recording medium interface 814, and a bus 815 connecting these components to one another are included.

  The CPU 801 controls the operation of each component of the topic providing server 4 ′, and processes of the statement analysis unit 43, the preference analysis unit 44, the topic analysis unit 45, and the information update unit 46 according to the present invention under control of the OS. Control the execution of

  The RAM 802 temporarily stores programs for executing the processes executed by the CPU 801 and data used during the execution of those programs. Further, as described above, the related word dictionary 50, the comparison result data 54 and the like may be stored. The ROM 803 stores programs and the like that are executed when the topic providing server 4 'starts up.

  The network interface 804 is an interface for connecting to the network 900. The network 900 is, for example, a network between the conversation control terminal apparatus 2 ′ ′ shown in FIG. 20 and a computer on which the crawler 730 operates, or a network such as the Internet.

  The audio control unit 805 controls the microphone 806 and the speaker 807 to control input / output of voice. The display controller 808 is a dedicated controller for actually processing a drawing command issued by the CPU 801. The display 809 is a display device configured of, for example, a liquid crystal display (LCD) or a cathode ray tube (CRT).

  The input device interface 810 receives a signal input from the keyboard 811 or the mouse 812 and transmits a predetermined command to the CPU 801 according to the signal pattern.

  The external storage device 813 is, for example, a storage device such as a hard disk or a semiconductor memory, and the above-described program and data are recorded in this device, and loaded from the RAM 802 as necessary during execution. For example, as described above, the related word dictionary 50, the comparison result data 54, and the like may also be stored.

  The external recording medium interface 814 accesses the external recording medium 910 to read data recorded therein. The external recording medium 910 is, for example, a portable flash memory, a CD (Compact Disc), a DVD (Digital Versatile Disc), or the like. A program that is executed by the CPU 801 to implement each function of the present invention can be provided from the external recording medium 910 via the external recording medium interface 814. Moreover, as another distribution form of the program for realizing each function of the present invention, a route is stored from the predetermined server on the network to the external storage device 813 or the RAM 802 through the network 900 and the network interface 804. Etc. are also conceivable.

  Although an example was demonstrated about the hardware constitutions of subject provision server 4 'which concerns on one Embodiment of this invention, the computer by which conversation control terminal device 2' 'and crawler 730 which are contained in the information retrieval system 100 of this invention operate | moves The hardware configuration is also basically the same as the configuration shown in FIG. However, here, with regard to the computer on which the topic providing server 4 ′ and the crawler 730 operate, the audio control unit 805, the microphone 806, the speaker 807, the display controller 808, the display 809, the input device interface 810, the keyboard 811 and the mouse 812 It is not a required component.

  In addition, the information retrieval system 100 described so far is controlled by the scenario data 28 (or the scenario data 55) between the Topic 20 operating on the conversation control terminal 2 ′ ′ and the topic providing server 4 ′. By transmitting and receiving data, information such as related words is displayed on the display of the conversation control terminal device 2 ''. In addition, for example, the Topic 20 is downloaded to the conversation control terminal 2 ′ ′ at a predetermined timing and activated, and the Topic 20 communicates with the topic providing server 4 ′ via a network such as the Internet, so-called cloud computing Can be configured as a system.

  However, the information search system 100 according to the present invention can effectively realize the technical idea of the present invention by various other configurations and methods. For example, each function of the topic providing server 4 'described above is configured by a WEB server, an ASP (Active Server Pages) server, etc., and a general WEB browser operating on the conversation control terminal device 2' 'controls scenario data. To display information such as related words or the like on the display of the conversation control terminal 2 ′ ′ according to the user's instruction without being The comparison result data can be configured to be switched. In this case, the WEB server, the ASP server, etc., functioning as the topic providing server 4 ', display data (for example, HTML data) for displaying on the display of the conversation control terminal device 2' '. Edit and generate.

<<< Other system configuration of topic offer system >>>
Next, with reference to FIG. 51, an outline of another system configuration of the topic providing system will be described.

  The topic providing system 1 ′ illustrated in FIG. 51 includes a conversation control terminal device 1002 (Topiclet 1020), a topic providing server 1004 (iWA 1030), a maintenance device 1003 (iWA Manager 1040), and a topic analysis device 1005. In the topic providing system 1 shown in FIG. 4, provision of the result (to the scenario data editing unit 410) by the topic analysis processing is shown to be performed in the maintenance device 3 (iWA Manager 40), but it is shown in FIG. In the topic providing system 1 ′, it is shown that the topic analysis processing itself is executed by a device separate from the maintenance device 1003 (iWA Manager 1040), which is the topic analysis device 1005, and the topic providing system is configured as such a system configuration. It can also be realized by

  The topic providing system 1 and the topic providing system 1 ′ are similar to the other configurations, and the detailed description will be omitted. The conversation control terminal device 1002 includes a receiving unit 1240 and a transmitting unit 1230. The receiver 1240 corresponds to the receiver 240 of the conversation control terminal device 2, and the transmitter 1230 corresponds to the transmitter 230 of the conversation control terminal device 2. The conversation control terminal device 1002 is basically the same as the conversation control terminal device 2 of FIG. 4 and the display of the other components is omitted. The topic providing server 1004 includes an input information analysis unit 1310 and a scenario data storage unit 1320. The maintenance device 1003 includes a scenario data transmission unit 1430, a scenario data editing unit 1410, and a terminal device virtual construction unit 1420. The topic analysis device 1005 includes a topic analysis unit 1510.

  In the topic providing system 1 ′ shown in FIG. 51, the topic analysis unit 1510 of the topic analysis device 1005 is connected to the topic providing server 1004 via a network, and the scenario data and other data obtained by the topic analysis process are , And provides the input information analysis unit 1310 of the topic providing server 1004. Also, the topic analysis unit 1510 of the topic analysis device 1005 is connected to the maintenance device 1003 via the network (or indirectly connected via the topic providing server 1004), and the scenario obtained by the topic analysis processing Data and other data are provided to the scenario data editing unit 1410 of the maintenance device 1003.

  The topic analysis unit 1510 of the topic analysis apparatus 1005 generates a topic list, and edits and verifies scenario data based on the topic list. The topic list is data in which the approach and connection of topics are given through the related words that relate topics. The topic analysis unit 1510 accumulates related words associated with topics in the topic list. The topic list provided to the maintenance device 1003 is data provided to the contractor of the topic providing system 1 ′, and these data are used, for example, when generating scenario data.

  In addition, the topic analysis unit 1510 of the topic analysis device 1005 may be configured to implement the processing of the sentence analysis unit 43 of the topic providing server 4 ′ shown in FIG. It may be configured to be implemented.

  Furthermore, a plurality of topic analysis units 1510 of the topic analysis device 1005 can be arranged to be associated with each of the subscribers of the topic providing system 1 '. In that case, the information acquired by each topic analysis device 1005 can be used (or used after being organized / integrated) and data can be provided to the corresponding maintenance device 1003 or topic provision server 1004. The timing at which data is provided by the topic analysis unit 1510 of the topic analysis apparatus 1005 and the content of the data may be different depending on the provision destination (that is, between the maintenance apparatus 1003 and the topic provision server 1004).

  Also, the contractor (or a person in charge belonging to the contractor) can use the maintenance device 1003 to construct a topic list or edit / create scenario data from the data provided from the topic analysis device 1005. However, each of the maintenance devices 1003 has a configuration such as a topic providing system 1 ′ shown in FIG. 51, and a mechanism such as Platform as a Service (PaaS) or Software as a Service (SaaS) that can be used remotely as a service on the Internet. It is also possible to realize the function.

  Up to this point, the information search system 100 according to an embodiment of the present invention has been described by showing some examples in which the present invention can be implemented, but these examples are for explaining the present invention. This is merely an example, and the scope of the present invention is not limited to these examples. The technical idea of the present invention can be realized by various methods and configurations other than these embodiments.

  This is a topic control system that can realize natural responses to the flow of topics and conversations, and can separately perform maintenance tasks such as changing, adding, and modifying topic control rules, so advanced knowledge This system can be used when there is a purpose of providing the user with important keywords extracted from topics or topics without having it.

1, 1 'topic provision system 2, 1002 conversation control terminal device (Topiclet 20, Topict 1020)
2 ', 2''conversation control terminal 3, 1003 maintenance device (iWA Manager 40, iWA Manager 1040)
4, 4 ', 1004 topic offer server (iWA30, iWA1030)
DESCRIPTION OF REFERENCE NUMERALS 10 topic storage device 21 input control unit 22 search control unit 23 transmission control unit 24 reception control unit 25 response information determination unit 26 output control unit 41 input information analysis unit 42 external log acquisition control unit 43 sentence analysis unit 44 preference analysis unit 45 topic Analysis unit 46 Information updating unit 100 Information retrieval system 1005 Topic analysis device

Claims (4)

An input information analysis unit that analyzes input information input from a user and generates input specific information;
A scenario data storing unit for storing the scenario data defining a process for output the response information about the topic,
A topic providing server that extracts scenario data from the scenario data storage unit based on the input identification information;
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the input identification information ;
And an output unit for outputting the response information determined by the pre-Symbol response information determining unit,
The state control index is an index related to at least one of the past input information and the past response information.
The response information determining unit, topic provides a system including a defined processed by the scenario data, by referring to the pre-Symbol state control index, and a conversation control terminal for determining the response information. The scenario data includes a plurality of statements each defining the process,
The response information determination unit
Based on the scenario data, the content of the state control index is determined by a predetermined statement of the scenario data, and from the predetermined statement to another statement according to the determination result of the state control index, the transition The topic providing system according to claim 1, wherein the response information according to the state control indicator is determined by executing a process corresponding to the previous statement. An input unit for the user to input input information;
A transmitter for transmitting the input information to the outside;
A receiving unit that receives scenario data defining processing for outputting input identification information generated by analyzing the transmitted input information and response information on a topic;
A scenario data storing unit for storing received said scenario data,
A state control indicator storage unit that stores state control indicators related to the input information and the response information;
A response information determination unit that determines the response information based on the scenario data and the input identification information ;
And an output unit for outputting the response information determined by the pre-Symbol response information determining unit,
The state control index is an index related to at least one of the past input information and the past response information.
The response information determining unit, by been processed defined in the scenario data, by referring to the state control index, the conversation control terminal for determining the response information. The scenario data includes a plurality of statements each defining the process,
The response information determination unit
Based on the scenario data, the content of the state control index is determined by a predetermined statement of the scenario data, and from the predetermined statement to another statement according to the determination result of the state control index, the transition The conversation control terminal device according to claim 3, wherein the response information corresponding to the state control index is determined by executing a process corresponding to the previous statement.

Images