JP2014106832A - Information providing apparatus and information providing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a user-friendly user interface that provides users with information.SOLUTION: An information providing apparatus 100 comprises: a data acquisition unit 31 for acquiring information data extracted in accordance with at least one retrieval key word; an extraction unit 32 for extracting key words related to the at least one retrieval key word from the information data acquired by the data acquisition unit 31; a message generation unit 33 for generating dialog message data including the related key words extracted by the extraction unit 32; and a data output unit 34 for outputting the dialog message data generated by the message generation unit 33.

Description

  The present invention relates to a user interface technique for providing information to a user.

  Currently, due to the spread of the Internet and the like, the information you want is available when you need it. For example, a smartphone user can immediately browse information related to a keyword by inputting a keyword of desired information and executing a search via the touch panel of the smartphone.

  The following Patent Document 1 proposes a chat system using the Internet. This system extracts a predetermined keyword included in the utterance sentence by performing morphological analysis on the utterance sentence acquired from the user terminal, and uses the extracted keyword to correspond to the keyword. A response sentence to be transmitted to the user terminal is determined using a conversation DB in which a plurality of response sentences are registered. When the predetermined keyword cannot be extracted from the remarks acquired from the user terminal, the system searches the keyword using the keyword search DB, and when the keyword cannot be extracted, the system transmits a question sentence to be transmitted to the user terminal. create. Further, the system collects information from the Web on the Internet based on a predetermined keyword, and registers the collected information in the conversation DB or the keyword search DB.

JP 2004-145541 A

  According to the above-described proposed method, a response sentence including information collected from the Internet related to a keyword included in a user's comment sentence acquired by voice input or character input is output at the user terminal. obtain. However, in this proposed method and normal Internet search, a large amount of information may be obtained depending on the keyword used for the search. In such a case, the user needs to find out desired information from the large amount of information.

  The present invention has been made in view of such circumstances, and provides a user-friendly user interface that provides information to a user.

  Each aspect of the present invention employs the following configurations in order to solve the above-described problems.

  A 1st aspect is related with an information provision apparatus. An information providing apparatus according to a first aspect relates to a data acquisition unit that acquires information data extracted according to at least one search keyword, and information related to the at least one search keyword from information data acquired by the data acquisition unit An extraction unit that extracts related keywords, a message generation unit that generates dialog message data including the related keywords extracted by the extraction unit, and a data output unit that outputs dialog message data generated by the message generation unit, Have.

  The second aspect relates to an information providing method. In the information providing method according to the second aspect, at least one computer acquires information data extracted according to at least one search keyword, and the related information related to the at least one search keyword is acquired from the acquired information data. Extracting a keyword, generating interactive message data including the extracted related keyword, and outputting the generated interactive message data.

  Another aspect of the present invention may be a program that causes at least one computer to implement the configuration of the first aspect, or a computer-readable recording medium that records such a program. May be. This recording medium includes a non-transitory tangible medium.

  According to each aspect described above, a user-friendly user interface that provides information to the user can be provided.

It is a figure which shows notionally the process structural example of the information provision apparatus which concerns on embodiment of this invention. It is a conceptual diagram which shows the structural example of the speech dialogue server apparatus in 1st Embodiment. It is a figure which shows notionally the process structural example of the speech dialogue server apparatus in 1st Embodiment, and a client apparatus. It is a flowchart which shows the operation example of the speech dialogue server apparatus in 1st Embodiment. It is a flowchart which shows a part of operation example of the speech dialogue server apparatus in 2nd Embodiment. It is a figure which shows notionally the process structural example of the speech dialogue server apparatus in 3rd Embodiment, and a client apparatus.

  Embodiments of the present invention will be described below. In addition, embodiment mentioned below is an illustration and this invention is not limited to the structure of the following embodiment.

  FIG. 1 is a diagram conceptually illustrating a processing configuration example of an information providing apparatus 100 according to an embodiment of the present invention. As shown in FIG. 1, the information providing apparatus 100 includes a data acquisition unit 31 that acquires information data extracted in accordance with at least one search keyword, and at least the information data acquired by the data acquisition unit 31. An extraction unit 32 that extracts a related keyword related to one search keyword, a message generation unit 33 that generates dialog message data including the related keyword extracted by the extraction unit 32, and an interactive message generated by the message generation unit 33 And a data output unit 34 for outputting data.

  The information providing apparatus 100 has, for example, a hardware configuration similar to that of the voice conversation server apparatus 1 in the detailed embodiment described later, and a program is processed in the same manner as the voice conversation server apparatus 1 so that each process described above is performed. Part is realized.

  The information providing method according to the embodiment of the present invention includes the following steps executed by at least one computer, such as the information providing apparatus 100 shown in FIG. That is, in the information providing method, at least one computer acquires information data extracted according to at least one search keyword, and a related keyword related to the at least one search keyword is obtained from the acquired information data. Extracting, generating interactive message data including the extracted related keywords, and outputting the generated interactive message data.

  In this embodiment, information data extracted from an information providing site, a database, or the like according to at least one search keyword is acquired. This information data may be information data including information content that matches the search keyword, or information data indicating a search result using the search keyword. For example, when information data is extracted from an information providing site such as an encyclopedia or a dictionary, or from a database, for example, the information data may indicate information content that matches the search keyword. When information data is acquired by a search function of an information providing site, for example, the information data may include an information item list (link list) acquired by the search.

  If the information data acquired in the present embodiment is data that is extracted according to at least one search keyword and includes related keywords related to the at least one search keyword, the extraction source and the data form thereof are Not limited. Here, the related keyword means a word or a word string that is not completely identical to the search keyword and has some relationship with the search keyword. The related keyword may be a character string partially including the search keyword. Further, when the search keyword is a person name, for example, the related keyword may be a word or a word string indicating one attribute of the person such as a birthday, a parent's name, a birth place, an occupation, and the like. Further, when the search keyword is “place name” and “food”, for example, the related keyword may be a name of a store or a special product that can be obtained by a search using the search keyword.

  In the present embodiment, such a related keyword is extracted from such information data, and dialogue message data including the related keyword is generated. The dialogue message indicated by this data is, for example, confirmation to the user or further information such as “Do you want to check (related keyword)?”, “Do you want to check (related keyword)?” This message prompts the request.

  Then, this dialog message data is output. There are various output forms such as output to a file, transmission to another computer, display on a display device, and printing on a printing device. However, this embodiment does not limit the output form of the dialog message data.

  Thereby, the user who has seen the dialogue message via the medium on which the dialogue message data is output can know the related keyword related to the information obtained using the search keyword. Even if the information obtained using the search keyword is not the expected information, the user can approach the expected information by using the related keyword. Furthermore, even if the information obtained using the search keyword is information that is initially expected, the user can receive a related topic by the conversation message, and if the user requests further information on the topic, Further information that was not scheduled can be obtained. Thus, according to the present embodiment, a user-friendly user interface that provides information to the user is realized.

  Hereinafter, the details of the above-described embodiment will be described. Below, 1st Embodiment, 2nd Embodiment, and 3rd Embodiment are illustrated as detailed embodiment. Each of the following detailed embodiments is an example when the above-described information providing apparatus 100 and the information providing method are applied to a cloud-type voice conversation server apparatus. However, the information providing apparatus 100 and the information providing method described above are not limited to application to a voice dialogue system that uses voice as an input / output form, and can be applied to apparatuses or systems of various input / output forms.

[First Embodiment]
〔Device configuration〕
FIG. 2 is a conceptual diagram illustrating a configuration example of the voice conversation server device (hereinafter simply referred to as a conversation server) 1 in the first embodiment. The dialogue server 1 is communicably connected to the client device 2 used by the user and the information providing site 5 via the communication network 3. The communication network 3 is a public network such as the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a wireless communication network, or the like. In the present embodiment, the communication mode among the dialogue server 1, the client device 2, and the information providing site 5 is not limited.

  The information providing site 5 is a general computer, and is a server device that provides various types of information such as a WEB server and a database server. The present embodiment does not limit the hardware configuration and software configuration of the information providing site 5.

  The dialogue server 1 obtains the user's comment text data from the client device 2 and searches the information providing site 5 based on the comment text data to provide the search result to the client device 2.

  As shown in FIG. 2, the dialogue server 1 includes a CPU (Central Processing Unit) 10, a memory 11, an input / output interface (I / F) 12, and a communication device that are mutually connected via a bus 15 as a hardware configuration. 13 etc. The memory 11 is a random access memory (RAM), a read only memory (ROM), a hard disk, a portable storage medium, or the like. The input / output I / F 12 is connected to a device that accepts an input of a user operation such as a keyboard and a mouse, and a device that provides information to the user such as a display device and a printer. The communication device 13 communicates with other nodes via the communication network 3. Note that the hardware configuration of the dialog server 1 is not limited.

  The client device 2 is a general user terminal such as a PC (Personal Computer), a tablet terminal, a smartphone, a mobile phone, or the like. Hereinafter, the client device 2 is simply referred to as a client 2. As shown in FIG. 2, the client 2 includes a CPU 10, a memory 11, an input / output I / F 12, a communication device 13, and the like that are connected to each other via a bus 15. The input / output I / F 12 includes an input device (not shown) that receives input of a user operation such as a keyboard and a mouse, a voice input unit 17 such as a microphone that acquires a user's speech, a speaker that outputs voice, and the like And an output device (not shown) for providing information to the user such as a display device and a printer. Note that the hardware configuration of the client 2 is not limited.

[Processing configuration]
FIG. 3 is a diagram conceptually illustrating a processing configuration example of the dialogue server 1 and the client 2 in the first embodiment. 3, the same processing units as those of the information providing apparatus 100 shown in FIG. 1 are denoted by the same reference numerals as those in FIG.

<Client device>
The client 2 includes a voice recognition unit 21, a voice synthesis unit 22, and the like. Each of these processing units is realized, for example, by executing a program stored in the memory 11 by the CPU 10. In addition, the program is installed from a portable recording medium such as a CD (Compact Disc) or a memory card or another computer on the network via the input / output I / F 12 or the communication device 13 and is stored in the memory 11. It may be stored.

  The speech recognition unit 21 generates speech text data corresponding to the speech speech by executing speech recognition processing on the speech speech of the user input and recorded through the speech input unit 17. The utterance text data is sent to the dialogue server 1 together with the identification information (noted as user ID) of the user as the output source. The comment text data is character data in which a voice uttered by the user is converted into text. In the present embodiment, a known technique may be used for the voice recognition process, and the voice recognition process itself and various voice recognition parameters used in the voice recognition process are not limited.

  The voice synthesis unit 22 performs voice synthesis processing on search result data, dialogue message data, and the like sent from the dialogue server 1, and outputs the search result and voice of the dialogue message via the voice output unit 18. The voice output from the voice output unit 18 is, for example, a voice that reads out information included in the search result data and dialogue message data. In the present embodiment, a known method may be used for the speech synthesis process, and the speech synthesis process itself is not limited.

<Voice dialogue server device>
As shown in FIG. 3, the dialogue server 1 includes a dialogue control engine 25, a WEB server processing unit 26, and the like. Each of these processing units is realized, for example, by executing a program stored in the memory 11 by the CPU 10. The program may be installed from a portable recording medium such as a CD or a memory card or another computer on the network via the input / output I / F 12 or the communication device 13 and stored in the memory 11. Good.

  The WEB server processing unit 26 realizes a general WEB server function. For example, the WEB server processing unit 26 enables bidirectional communication between the client 2 and the voice conversation server device 1, execution of a servlet on the WEB server, and the like. The present embodiment does not limit the specific processing of the WEB server processing unit 26.

  The dialog control engine 25 acquires the user's comment text data from the client 2, selects an appropriate information providing site 5 based on the comment text data, requests information from the information providing site 5, and the result To get. As shown in FIG. 3, the dialogue control engine 25 includes a speech data acquisition unit 35, a data acquisition unit 31, an extraction unit 32, a message generation unit 33, a data output unit 34, and the like. Similarly, each of these processing units is realized, for example, by executing a program stored in the memory 11 by the CPU 10. Hereinafter, the data acquisition unit 31, the extraction unit 32, the message generation unit 33, and the data output unit 34 will be described with a focus on the content different from the information providing apparatus 100 described above, and the same content will be omitted as appropriate.

  The utterance data acquisition unit 35 acquires the user's utterance text data and user ID sent from the client 2 via the WEB server processing unit 26. The user's message text data acquired by the message data acquisition unit 35 may indicate a response message returned from the client 2 with respect to the dialog message data or confirmation message data transmitted from the dialog server 1 to the client 2. . In this case, the message text data is expressed as response message data. In addition, the comment text data may indicate an information request from a user including at least one search keyword. In this case, the message text data is represented as request message data. Therefore, the utterance data acquisition unit 35 can also be called a response utterance data acquisition unit and a request utterance data acquisition unit.

  The data acquisition unit 31 selects an appropriate information providing site 5 based on the request utterance data acquired by the utterance data acquisition unit 35, and at least one of the information providing sites 5 included in the request utterance data. Request a search using a search keyword. The data acquisition unit 31 acquires information data extracted in response to the search request from the information providing site 5. The contents and data form of the acquired information data are as described above. In the present embodiment, the method for selecting the information providing site 5 based on the request message data is not limited. The information providing site 5 associated with the word data included in the request utterance data or the word data associated therewith may be selected. The information providing site 5 may be fixedly determined.

  The data acquisition unit 31 uses, as a search keyword, at least one word data out of a plurality of word data obtained by decomposing the requested utterance data. As search keywords, word data such as nouns, noun phrases, verbs, verb phrases and the like are used.

  Further, the data acquisition unit 31 acquires new information data related to the related keyword included in the dialogue message data in accordance with the response content indicated by the response message data acquired by the message data acquisition unit 35. The data acquisition unit 31 indicates the position information of the information data related to the related keyword (for example, URL (Uniform Resource Locator)) in the information data from which the related keyword is extracted. Information data related to the related keyword may be acquired from the information providing site 5. Further, when the location information of the information data related to the related keyword is not included, the data acquisition unit 31 may further request the information providing site 5 to perform a search using the related keyword as a search keyword. .

  The response content of the response message data is specified by analyzing word data obtained by decomposing the response message data. For example, whether or not the word data obtained by part-of-speech decomposition includes a positive predetermined word such as “Yes”, whether or not a negative predetermined word such as “No” is included, etc. The analysis determines whether the response content is positive or negative. Any processing unit may perform the analysis of the response content of the response message data.

  When the information data acquired by the data acquisition unit 31 has information content that matches the search keyword, the extraction unit 32 extracts text data indicating the information content that matches the search keyword included in the information data To do. Here, the information content matching the search keyword means the information content corresponding to one item of the information item list, unlike the information item list indicating the search result by the information providing site 5. Further, the information content that matches the search keyword is not only information that completely matches the search keyword, but also the information content of one information item that is estimated to match the search keyword and extracted by the information providing site 5. Good.

  For example, when information data provided from the information providing site 5 which is a dictionary site is acquired by a search using the search keyword “Sendai Castle”, the explanatory text data about “Sendai Castle” is obtained from the information data. Is acquired. When the information data has information content that matches the search keyword, the extraction unit 32 extracts text data indicating the information content that matches the search keyword included in the information data.

  The extraction unit 32 further extracts a related keyword related to at least one search keyword included in the request message data from the information data acquired by the data acquisition unit 31. According to the above-described example, “Date Masamune”, “Age Castle Year”, “Abandoned Castle Year”, and the like are acquired as related keywords included in the information data. The extraction unit 32 may extract a plurality of related keywords when the information data includes a plurality of related keywords. Moreover, the extraction part 32 should just extract only a related keyword, when information data does not have the information content which corresponds to the said search keyword.

  In addition, the extraction unit 32 determines the related keyword included in the transmitted dialogue message from the information data acquired by the data acquisition unit 31 according to the response content indicated by the response message data acquired by the message data acquisition unit 35. The text data indicating the information content is further extracted.

  The message generation unit 33 generates dialog message data including the related keyword extracted by the extraction unit 32, and generates information providing message data including text data indicating information content that matches the search keyword extracted by the extraction unit 32. Generate more. The message generation unit 33 does not generate information provision message data when the extraction unit 32 does not extract text data indicating the information content. For example, the dialog message data is generated as text data indicating a question sentence for confirming whether information on a related keyword is expected. In addition, the dialog message data may be generated as text data indicating a question sentence asking whether or not information on related keywords is desired.

  When a plurality of related keywords are extracted by the extraction unit 32, the message generation unit 33 generates dialog message data including one related keyword selected from the plurality of extracted related keywords. The message generation unit 33 may include a plurality of related keywords selected from the plurality of related keywords extracted by the extraction unit 32 in the dialog message data. In this case, the message generation unit 33 holds information that makes it possible to identify a related keyword already included in the dialogue message data and other related keywords among the plurality of related keywords extracted by the extraction unit 32. . The message generation unit 33 refers to the held information according to the response content indicated by the response message data received from the client 2 after transmitting the conversation message data. The message generation unit 33 selects one new related keyword that has not been selected last time from a plurality of related keywords based on the stored information, and newly includes the selected one new related keyword. Such interactive message data may be generated.

  When the extraction unit 32 extracts text data indicating the information content of the related keyword, the message generation unit 33 generates information provision message data including the text data.

  The data output unit 34 transmits the above-described information providing message data to the client 2 that is the transmission source of the request message data, and then transmits the above-described dialog message data to the client 2. The information providing message data and the dialogue message data may be sequentially transmitted with a predetermined time interval, or may be transmitted in a batch in a state of being combined into one data. In the case of batch transmission, the information providing message data and the conversation message data may be combined into one data so that each can be identified.

  The data output unit 34 further transmits the new dialog message data generated by the message generation unit 33 based on the response message data transmitted from the client 2 and received by the message data acquisition unit 35 after transmitting the dialog message data. 2 to send. Further, when the information providing message data is not generated, the data output unit 34 transmits only the dialog message data to the client 2.

[Operation example]
Hereinafter, the information providing method in the first embodiment will be described with reference to FIG. FIG. 4 is a flowchart showing an operation example of the dialogue server 1 in the first embodiment.

  The dialogue server 1 acquires message text data from the client 2 (S41). The comment text data is generated, for example, by recording and recognizing the speech voice uttered by the user of the client 2 by the client 2, and is transmitted from the client 2 to the dialogue server 1. Here, the message text data is request message data indicating an information request from a user including at least one search keyword.

  The dialogue server 1 requests a certain information providing site 5 to perform a search using at least one search keyword included in the requested utterance data, and uses the information data extracted in response to the search request as the information providing site. 5 (S42).

  The dialogue server 1 determines whether or not the information content matching the search keyword included in the request message data acquired in (S41) is indicated in the information data acquired in (S42) (S43). When the information content matching the search keyword is indicated (S43; YES), the dialogue server 1 extracts text data indicating the information content, and generates information providing message data including this text data ( S44).

  The dialogue server 1 transmits the information providing message data to the client 2 that is the transmission source of the request message data acquired in (S41) (S45). Accordingly, the client 2 performs voice synthesis processing on the information provision message data, and outputs the voice of the information provision message via the voice output unit 18. As a result, the user can listen to the voice in which the retrieved information is read out according to the request message data. On the other hand, when the information content matching the search keyword is not indicated (S43; NO), the dialogue server 1 does not generate and transmit the information providing message data. Here, the dialogue server 1 deletes the already provided information providing message data.

  The dialogue server 1 extracts a related keyword related to the search keyword included in the request message data acquired in (S41) from the information data acquired in (S42) (S46). The dialogue server 1 generates dialogue message data including the extracted related keyword (S47). Here, the dialogue server 1 holds the selection information of the related keywords, that is, information that can identify the related keywords selected for inclusion in the dialogue message data and other related keywords (S48).

  Subsequently, the dialogue server 1 transmits dialogue message data to the client 2 (S50). Thus, the client 2 performs voice synthesis processing on the dialog message data, and outputs the voice of the dialog message via the voice output unit 18. As a result, the user can hear a sound confirming whether he or she expects information about the related keyword.

  Next, the dialogue server 1 acquires response message data sent back from the client 2 in response to the dialogue message data transmitted in (S50) and output as voice by the client 2 (S51). The response utterance data indicates the content of the response uttered by the user according to the voice corresponding to the dialogue message data, and is generated by recording and voice recognition of the user's utterance by the client 2. To the dialogue server 1. The response message data indicates, for example, an affirmative response or a negative response.

  The dialog server 1 determines whether or not the response message data further requests information on the related keyword included in the dialog message data transmitted in (S50) (S52). When the conversation server 1 indicates that the response message data acquired in (S51) further requests information about the related keyword (S52; YES), whether or not information related to the related keyword has already been acquired. Is determined (S53). When the information content about the related keyword is not indicated in the information data acquired in (S42) (S53; NO), the dialogue server 1 acquires new information data regarding the related keyword (S54). As described above, the dialog server 1 executes (S43) and subsequent steps on the acquired new information data.

  On the other hand, when the information content about the relevant keyword is indicated in the information data acquired in (S42) (S53; YES), the conversation server 1 determines from the information data acquired in (S42). Text data indicating the information content of the related keyword is extracted, and information providing message data including the text data is generated (S55). The dialogue server 1 transmits the information providing message data to the client 2 (S56). Thereby, the user can hear the voice which reads the information content about a related keyword.

  On the other hand, the dialogue server 1 indicates that the response message data acquired in (S51) indicates that the information about the related keyword is not further requested (S52; YES), or the related keyword as described above. After transmitting the information provision message data about, it is determined whether or not there is an unselected related keyword among the related keywords extracted in (S45) based on the information held in (S47) ( S57). When there is an unselected related keyword (S57; YES), the dialog server 1 selects a new related keyword from the unselected related keywords (S58), and a new dialog including the new related keyword. Message data is generated (S47).

  Thereafter, the dialog server 1 executes (S48) and subsequent steps as described above based on the new dialog message data. If there is no unselected related keyword (S57; NO), the dialogue server 1 ends the process.

  The information providing method in the present embodiment is not limited to the execution order of the steps shown in FIG. For example, the transmission of information providing message data (S45) and the transmission of dialog message data (S50) are not limited immediately after the generation of each message. For example, after the transmission of the information providing message data (S45), the elapse of a predetermined time may be measured, and the transmission of the dialog message data (S50) may be performed after the elapse of the predetermined time. For the convenience of explanation, the acquisition of the comment text data is shown separately in (S41) and (S51), which are shown in one step, and in the subsequent steps, the content determination of the comment text data is performed. Then, the subsequent execution contents may be switched.

[Operation and Effect of First Embodiment]
As described above, in the first embodiment, the request utterance data corresponding to the utterance indicating the information request by the user is acquired, the search using the search keyword included in the request utterance data is requested to the information providing site 5, Information data corresponding to the search result is acquired. In general, a speech (conversation) that a user utters in order to obtain desired information is likely to include a word related to the desired information. Can be acquired appropriately.

  In the first embodiment, when information content matching the search keyword is indicated in the information data, an information providing message including text data indicating the information content is sent to the client 2, and the information content is Sound is output. Therefore, the user can hear the information retrieved in such a manner by voice in response to a statement indicating an information request. Thereby, according to 1st Embodiment, the information provision service by the natural dialog between a user and the dialog server 1 is realizable.

  Furthermore, in the first embodiment, a related keyword related to the search keyword is extracted from the information data acquired from the information providing site 5, and dialog message data including the related keyword is sent to the client 2, and the dialog message Is output as audio. Thereby, the user can know the related keyword related to the information obtained by using the search keyword.

  Furthermore, in the first embodiment, response message data indicating the response content spoken by the user is acquired as a response to the dialogue message data, and the related keyword included in the dialogue message data according to the response message data Information about is obtained. Information related to the related keyword is extracted from information data already acquired according to the search keyword, information data newly acquired from the information providing site 5, and the like. And the information content regarding the related keyword extracted in this way is output by voice.

  Thereby, if the user utters a response to the dialogue message that is output as voice, the user can further acquire information on the related keyword included in the dialogue message. Therefore, according to the first embodiment, it is possible to easily bring the user close to expected information by sending the dialogue message data including the related keyword to the client 2. Further, according to the first embodiment, the conversation message data can provide the user with a related topic and allow the user to acquire further information that was not initially scheduled. According to the first embodiment, it is possible to realize an information providing service for performing expansive user interaction while providing information desired by a user.

  Furthermore, in the first embodiment, when a plurality of related keywords are included in the information data acquired in accordance with the previous request message data, dialogue message data including each related keyword is sequentially transmitted. Accordingly, the user can easily acquire further information by giving a positive response only to the interactive message including the related keyword desired by the user. Thus, according to the present embodiment, a user-friendly user interface that provides information to the user can be realized.

[Second Embodiment]
Information content included in information data obtained from the information providing site 5 may be long. In such a case, all of the text data indicating the information content that matches the search keyword may be included in the information providing message data, or the partial text data may be included in the information providing message data. In the second embodiment, when the information content is a long sentence, the partial text data is included in the information providing message data. Hereinafter, the content different from the first embodiment will be mainly described for the dialogue server 1 in the second embodiment. In the following description, the same contents as those in the first embodiment are omitted as appropriate.

[Processing configuration]
The processing configuration of the dialog server 1 in the second embodiment may be the same as that of the first embodiment shown in FIG. However, in the second embodiment, the processing content of each processing unit shown below is different from that of the first embodiment.

  The message generation unit 33 generates information providing message data in which text data indicating information content that matches the search keyword extracted by the extraction unit 32 includes partial text data separated at a predetermined position. The predetermined position is specified using, for example, a predetermined number of punctuation marks. For example, the message generator 33 holds information about two punctuation marks in advance, and extracts the partial text data by dividing the text data at the position of the second punctuation mark from the beginning of the sentence.

  At this time, the message generator 33 further generates confirmation message data for confirming whether or not to continue providing information. The confirmation message data is formed of text data indicating a sentence for confirming whether the user wants to hear the information content indicated by the next partial text data.

  The data output unit 34 transmits the information provision message data including the partial text data and the confirmation message data to the client 2.

[Operation example]
Hereinafter, an information providing method according to the second embodiment will be described with reference to FIG. FIG. 5 is a flowchart showing a part of an operation example of the dialogue server 1 in the second embodiment. FIG. 5 shows a partial operation example from (S42) to (S46) shown in FIG. Steps not shown in FIG. 5 are performed as shown in FIG.

  The dialogue server 1 determines whether or not the information content matching the search keyword included in the request message data acquired in (S41) is indicated in the information data acquired in (S42) (S43). If the information content matching the search keyword is indicated (S43; YES), the dialogue server 1 extracts text data indicating the information content (S44). The operation of the second embodiment so far is the same as that of the first embodiment.

  The dialogue server 1 investigates a predetermined position related to the extracted text data (S61). Specifically, the dialogue server 1 determines whether or not the predetermined position exists in the text data (S61). This corresponds to the determination of whether or not the text data is a long sentence that should be intermittently output as a voice. When the predetermined position does not exist in the text data (S61; NO), the dialogue server 1 generates information provision message data including all the text data extracted in (S44) (S68).

  Subsequently, the dialogue server 1 transmits the information providing message data to the client 2 that is the transmission source of the request message data acquired in (S41) (S69). As a result, the client 2 performs voice synthesis processing on the information providing message data, and outputs a voice for reading out all information contents indicated by the information data acquired from the information providing site 5 via the voice output unit 18.

  On the other hand, when the predetermined position exists in the text data (S61; YES), the dialogue server 1 extracts partial text data in which the text data is divided at the predetermined position (S62). At this time, the dialogue server 1 holds the remaining text data obtained by removing the partial text data extracted in (S62) from all the text data extracted in (S44). The dialogue server 1 generates information provision message data including the partial text data extracted in (S62) (S63).

  The dialogue server 1 further generates confirmation message data for confirming the necessity of continuing information provision (S64).

  The dialogue server 1 uses the information providing message data including the partial text data generated in (S63) and the confirmation message data generated in (S64) as the transmission source of the request message data acquired in (S41). To the client 2 (S65). As a result, the client 2 performs voice synthesis processing on the information providing message data, and outputs a voice that reads out a part of the information content indicated by the partial text data via the voice output unit 18. Further, the client 2 performs voice synthesis processing on the confirmation message data, and outputs a voice for confirming whether or not it is necessary to continue providing information via the voice output unit 18.

  Next, the dialogue server 1 acquires response message data returned from the client 2 in response to the confirmation message data (S66). The response utterance data indicates the content of the response uttered by the user in response to the confirmation voice corresponding to the confirmation message data, and is generated by recording and voice recognition of the user's utterance in the client 2. 2 to the dialogue server 1. The response message data indicates, for example, an affirmative response or a negative response.

  The dialogue server 1 determines whether or not the response message data is a request for continuation of information provision (S67). When the response utterance data acquired in (S66) indicates that the continuation of information provision is requested (S67; YES), the dialogue server 1 executes the subsequent steps (S61) on the remaining text data to be held. . When the predetermined position exists in the remaining text data to be held (S61; YES), the dialogue server 1 extracts partial text data from the remaining text data (S62), and based on this partial text data As described above, (S63) and subsequent steps are executed.

  On the other hand, when the predetermined position does not exist in the remaining text data to be held (S61; NO), the dialogue server 1 generates information providing message data including the remaining text data (S68), and the information The provided message data is transmitted (S69). In this case, since all of the text data extracted in (S44) is intermittently transmitted in the information providing message data, the dialog server 1 executes (S46) and subsequent steps. To do.

  The dialogue server 1 executes the processing from (S46) onward even when the response message data acquired in (S66) does not require continuation of information provision (S67; NO).

[Operation and Effect of Second Embodiment]
In the second embodiment, when a predetermined position exists in the text data indicating the information content obtained by the search from the information providing site 5, the information providing message data including a part of the text data, the information providing message Confirmation message data for confirming the necessity of continuation is generated and transmitted to the client 2. As a result, a part of the information content obtained as a search result is output as a sound, and then a confirmation sound indicating whether it is necessary to continue providing information is output. And according to the statement which asks for continuation by a user, the information contents concerned are intermittently provided to a user.

  As described above, according to the second embodiment, when long text information is acquired as a search result, the information is intermittently output while confirming the user's intention. You can hear only the sound you want to hear without being asked. Further, even when listening to the voice of all information contents, the user can secure a necessary time in the middle, such as a time for taking notes. Therefore, according to the second embodiment, an information providing service through a more natural dialogue can be realized.

[Third Embodiment]
In each of the above-described embodiments, the request utterance data indicating the same request content by the same user may be acquired by the dialogue server 1 a plurality of times. At this time, if an interactive message including the same related keyword is output every time, the user may feel bothered or bored. Therefore, in the third embodiment, when request utterance data indicating the same request content by the same user is acquired a plurality of times within a predetermined period, an interactive message including different related keywords is output each time. Hereinafter, the content different from the first embodiment and the second embodiment will be mainly described for the dialogue server 1 in the third embodiment. In the following description, the same contents as those in the first embodiment and the second embodiment are omitted as appropriate.

[Processing configuration]
FIG. 6 is a diagram conceptually illustrating a processing configuration example of the dialogue server 1 and the client 2 in the third embodiment. The dialogue server 1 in the third embodiment further includes a request holding unit 61 in addition to the first embodiment and the second embodiment. The request holding unit 61 is realized on the memory 11, for example. In the present embodiment, for convenience of explanation, the request holding unit 61 is shown separately from other processing units. However, the request holding unit 61 may be realized as an information holding unit inside another processing unit. Good.

  The request holding unit 61 holds information regarding the request message data acquired by the message data acquisition unit 35 for each user and for each request content. Specifically, the request holding unit 61 holds the user ID, the information providing site 5, the search keyword, the selection status of related keywords, the reception time, etc., for each request message data. As the user ID, the user ID acquired by the message data acquisition unit 35 together with the request message data is held as information indicating the transmission source. As the information providing site 5, information specifying the information providing site 5 selected as the request destination by the data acquisition unit 31 based on the request message data is held. As the search keyword, the search keyword extracted from the request message data and sent together with the information request to the requested information providing site 5 is held. As related keyword selection information, among a plurality of related keywords extracted from information data obtained as a search result from the information providing site 5, related keywords selected for inclusion in the dialog message data and other related keywords Is stored. As the reception time, the reception time of the request message data is held.

  The request holding unit 61 monitors the elapsed time from the reception time of each record, and deletes the record corresponding to the request message data whose elapsed time has exceeded a predetermined time. Thus, in this embodiment, the request holding unit 61 holds the user ID and monitors the elapsed time from the reception time. However, the dialog control engine 25 is realized by a processing unit such as a thread or a process that is generated for each user (or client 2) in response to reception of the request message data and ends after a predetermined time has elapsed since the reception. In this case, it is not necessary to monitor the user ID and elapsed time in the request holding unit 61.

  When the data acquisition unit 31 selects the information providing site 5 and extracts the search keyword based on the request message data, whether the request message data and the request message data indicating the same request content by the same user have already been acquired. It is determined by referring to the request holding unit 61. Specifically, when the data acquisition unit 31 selects the information providing site 5 and extracts the search keyword based on the request message data acquired this time, the request holding unit 61 selects the information providing site 5 and the search keyword. In addition, the existence of a record holding the user ID is confirmed. When the corresponding record exists, the data acquisition unit 31 determines that request utterance data indicating the same request content by the same user has already been acquired.

  When the corresponding record does not exist, the data acquisition unit 31 causes the request holding unit 61 to hold the selected information providing site 5, the extracted search keyword, the user ID, and the reception time regarding the request message data acquired this time. In this case, the data acquisition unit 31 requests the information providing site 5 to perform a search using the search keyword, and retrieves the information data extracted in response to the search request as in the above embodiments. Obtain from the providing site 5. The data acquisition unit 31 may make a new search request when a corresponding record exists, that is, when request message data indicating the same request content by the same user has already been acquired. Information data that has already been acquired may be used without making a request.

  Based on the information held in the request holding unit 61 when the message data acquisition unit 35 acquires request message data for the second and subsequent times indicating the same request content from the same user, Generate interactive message data containing relevant keywords for selection. The message generation unit 33 uses the request holding unit 61 to store information that can identify a related keyword already included in the dialogue message data and other related keywords among the plurality of related keywords extracted by the extraction unit 32. Reflect in the corresponding record.

[Operation example]
Hereinafter, an information providing method according to the third embodiment will be described with reference to FIG.
In the third embodiment, the dialogue server 1 uses the information held in the request holding unit 61 when the second and subsequent request message data indicating the same request content from the same user is acquired in (S41). Based on this, an unselected related keyword is selected, and dialogue message data including the selected related keyword is generated (S47).

  In the third embodiment, before (S42), the request holding unit 61 determines whether the request message data acquired in (S41) is the second or later request message data indicating the same request content from the same user. Judge by reference. When the second and subsequent request message data is acquired in (S41), the dialogue server 1 may execute (S42), or may not execute (S42) and already execute based on the previous request message data. You may make it use the acquired information data. Similarly, when the information data already acquired based on the previous request message data is used without executing (S42), the dialogue server 1 selects the unselected keywords from the plurality of related keywords already extracted. Related keywords may be extracted.

  Each example will be listed below, and each of the above-described embodiments will be described in more detail. However, the present invention is not limited to the following examples. Each embodiment will be described below based on the flowchart of FIG.

  In the first embodiment, a case where the user of the client 2 remarks “Tell me about Leonardo da Vinci” is exemplified. In this case, the dialogue server 1 acquires request message data including text data “Tell me about Leonardo da Vinci” from the client 2 (S41). The dialogue server 1 selects the encyclopedia site as the information providing site 5 in accordance with the requested utterance data, and extracts the search keyword “Leonardo da Vinci”. The dialogue server 1 requests the encyclopedia site to perform a search using the search keyword, and obtains a WEB page including contents describing “Leonardo da Vinci” as information data (S42).

  The dialogue server 1 extracts the information content about “Leonardo da Vinci” from the WEB page, generates an information providing message including text data indicating the information content (S44), and sends the information providing message to the client 2 (S45). At this time, summary information in the acquired WEB page may be included in the information providing message.

  The dialogue server 1 extracts related keywords from the WEB page (S46). For example, the dialogue server 1 extracts a character string set by a <TH> tag in the WEB page as a related keyword. For example, a plurality of related keywords of “real name”, “birth”, “death”, and “representative work” are extracted. The dialogue server 1 selects one related keyword (for example, “representative work”) from the extracted plurality of related keywords, and generates dialogue message data including the related keyword (S47). For example, the dialogue message data is text data “Do you want to hear about the masterpiece?”.

  The dialogue server 1 displays information indicating that “representative work” has already been selected and the rest have not been selected among the extracted related keywords “real name”, “birth”, “dead”, and “representative work”. Hold (S48). The dialogue server 1 transmits the dialogue message data to the client 2 (S50).

  Here, it is assumed that the user has said “Yes, please” in response to the dialogue message. In this case, the response message data including the text data of the message is sent from the client 2 to the dialog server 1, and the dialog server 1 acquires the response message data (S51). The dialog server 1 has a positive response content of the response message data (S52; YES), and information related to the related keyword “masterpiece” is shown on the WEB page related to “Leonardo da Vinci”. (S53; YES), text data explaining the related keyword “representative work” is extracted from the WEB page, and information providing message data including the text data is generated (S55). As text data describing the related keyword “representative work”, for example, a character string set by a <TD> tag is extracted.

  The dialog server 1 transmits information providing message data related to the related keyword “representative work” to the client 2 (S56). Thereafter, since there is an unselected related keyword (S57; YES), the dialog server 1 next selects, for example, a new related keyword “birth” (S58). Thereafter, dialogue message data including the new related keyword “birth” is generated (S47) and transmitted to the client 2 (S50).

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “Tell me about Leonardo da Vinci”
Audio output: “Leonardo da Vinci is ... (explanation)”
Audio output: “Do you want to hear about the masterpiece?”
User: “Yes, please”
Audio output: “The main work is Mona Lisa, ... (explanation)”
Audio output: “Do you want to hear about birth?”

  On the other hand, it is assumed that the user says “No, it is fine” in response to the dialogue message “Do you want to hear about the masterpiece?” In this case, since the response content of the response message data is negative (S52; NO), the dialogue server 1 confirms the presence of an unselected related keyword as described above (S57). Thereafter, after (S58) is executed, (S47) and subsequent steps are executed.

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “Tell me about Leonardo da Vinci”
Audio output: “Leonardo da Vinci is ... (explanation)”
Audio output: “Do you want to hear about the masterpiece?”
User: “No, it ’s fine”
Audio output: “Do you want to hear about birth?”

  As described above, according to the first embodiment, information on “Leonardo da Vinci” originally expected by the user can be provided, and topics related to the information (“representative work”, “birth”, etc.) can be provided. ) Can be further provided. Thereby, the user can acquire further information that was not initially scheduled. Further, according to the first embodiment, even when a keyword about information originally expected by the user cannot be recalled, the information to be provided can be brought close to the intention of the user. Even if the user forgets the word “Mona Lisa”, which is originally expected information, and only the keyword “Leonardo da Vinci” is recalled, as described above, the user can use the related keyword “ You can get information on "Mona Lisa" from "Representative Works".

  In the second embodiment, a case where the user of the client 2 remarks “Tell me about Sendai” is exemplified. In this case, the dialogue server 1 acquires request message data including the text data “Tell me about Sendai” from the client 2 (S41). The dialogue server 1 selects the encyclopedia site as the information providing site 5 according to the request message data, and extracts the search keyword “Sendai”. When the dialog server 1 requests the encyclopedia site to perform a search using the search keyword, the search keyword “Sendai” is ambiguous, so the information item list that does not include the information content that matches “Sendai” is displayed. The included WEB page is acquired as information data (S42). For example, when Wikipedia (registered trademark) is selected as the encyclopedia site, and the character string “Wikipedia (registered trademark): ambiguity avoidance” is included in the acquired WEB page, the dialogue server 1 It is determined that the WEB page does not show information content that matches the search keyword (S43; NO).

  In this case, the dialogue server 1 extracts the related keyword from the WEB page without generating the information provision message data (S46). For example, the dialogue server 1 extracts a character string set by a <LI> tag in the WEB page as a related keyword. For example, a plurality of related keywords such as “Sendai City”, “Miyagi Prefecture”, “Sendai Station”, “Sendai Airport”, and “Sendai Plain” are extracted. The dialogue server 1 selects one related keyword (for example, “Sendai Station”) from among the plurality of extracted related keywords, and generates dialogue message data including the related keyword (S47). For example, the dialogue message data is text data “Do you mean Sendai Station?”.

  The dialogue server 1 has already selected “Sendai Station” among the extracted related keywords “Sendai City”, “Miyagi Prefecture”, “Sendai Station”, “Sendai Airport”, “Sendai Plain”, etc. Information indicating that the rest is not selected is held (S48). The dialogue server 1 transmits the dialogue message data to the client 2 (S50).

  Here, it is assumed that the user has said "Yes, yes" to the dialogue message. In this case, the dialogue server 1 has a positive response content of the response message data (S52; YES), and information related to the related keyword “Sendai Station” is not shown on the WEB page related to “Sendai”. (S53; NO), new information data related to the related keyword “Sendai Station” is acquired (S54). At this time, when the URL to the WEB page related to the related keyword “Sendai Station” is included in the WEB page related to “Sendai”, the dialogue server 1 acquires information data related to “Sendai Station” from the URL.

  The dialogue server 1 extracts the information content about “Sendai Station” from the WEB page, generates an information providing message including text data indicating the information content (S44), and transmits the information providing message to the client 2. (S45). Thereafter, (S46) and subsequent steps are executed for the WEB page acquired for "Sendai Station". For example, a related keyword such as “location” is extracted from the WEB page (S46).

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “Tell me about Sendai”
Audio output: “Do you mean Sendai Station?”
User: “Yes, yes”
Audio output: “Sendai station is ... (explanation)”
Audio output: “Do you want to ask about the location?”

  On the other hand, it is assumed that the user says “No, not” to the dialogue message “Do you mean Sendai Station?”. In this case, since the response content of the response message data is negative (S52; NO), the dialogue server 1 confirms the presence of an unselected related keyword as described above (S57). Thereafter, after (S58) is executed, (S47) and subsequent steps are executed.

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “Tell me about Sendai”
Audio output: “Do you mean Sendai Station?”
User: “No, not”
Audio output: “Do you mean Miyagi Prefecture?”
User: “No, not”
Audio output: “Do you mean Sendai City?”
User: “Yes, yes”
Audio output: "Sendai City is ... (explanation)"
Audio output: “Do you want to ask about the total population?”

  As described above, according to the second embodiment, an interactive message including a related keyword is provided even when the user cannot speak including an appropriate keyword and makes a comment including an ambiguous keyword (“Sendai”). Thus, specific item information to be included in the keyword can be provided to the user.

  In the third embodiment, a case where the user of the client 2 says, “I want to eat beef tongue in Sendai” is exemplified. In this case, the dialogue server 1 acquires request message data including text data “I want to eat beef tongue in Sendai” from the client 2 (S41). The dialogue server 1 selects the gourmet site as the information providing site 5 in accordance with the request message data, and extracts the search keywords “Sendai” and “Beef tongue”. When the dialogue server 1 requests a search using the search keyword for the gourmet site, the dialog server 1 acquires a WEB page including a plurality of store information lists hit by the search keyword as information data (S42). Since the gourmet site is selected as the information providing site 5, the dialogue server 1 determines that the first acquired WEB page does not show the information content that matches the search keyword (S43; NO).

  In this case, the dialogue server 1 extracts the related keyword from the WEB page without generating the information provision message data (S46). For example, the dialogue server 1 extracts a character string set by a <LI> tag in the WEB page as a related keyword. For example, a plurality of store names are extracted as related keywords. The dialog server 1 selects one store name from among the extracted store names, and generates dialog message data including the store name as a related keyword (S47). For example, the dialogue message data is text data “It seems that a shop called“ ... is popular ”. The dialogue server 1 retains information that can identify a store name that has already been selected from the extracted store names and other store names (S48). The dialogue server 1 transmits the dialogue message data to the client 2 (S50).

  Here, it is assumed that the user has remarked “the store is good” in response to the dialogue message. In this case, since the response content of the response message data is affirmative (S52; YES), the dialog server 1 does not display information on the store name selected as the related keyword on the WEB page (S53). NO), new information data related to the related keyword (store name) is acquired (S54). At this time, when the URL to the WEB page related to the store name selected as the related keyword is included in the WEB page as the search result, the dialogue server 1 acquires information data related to the store name from the URL.

  The dialogue server 1 extracts the information content about the store name from the WEB page, generates an information provision message including text data indicating the information content (S44), and transmits the information provision message to the client 2 (S45). ). Thereafter, (S46) and subsequent steps are executed for the WEB page acquired for the store name.

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “I want to eat beef tongue in Sendai”
Audio output: “... seems to be a popular shop”
User: “That store, like it”
Audio output: "... is ... (explanation)"

  On the other hand, it is assumed that the user has remarked that “a different store is good” in response to the dialogue message “... a store seems to be popular”. In this case, since the response content of the response message data is negative (S52; NO), the dialogue server 1 confirms the presence of an unselected related keyword as described above (S57). Thereafter, after (S58) is executed, (S47) and subsequent steps are executed.

As a result, the following dialogue is performed between the user and the dialogue server 1 via the client 2.
User: “I want to eat beef tongue in Sendai”
Audio output: “... seems to be a popular shop”
User: “I want a different store”
Audio output: “... is also a popular store”
User: “That store, like it”
Audio output: "... is ... (explanation)"

  As shown in Example 3, each of the above-described embodiments is not effective only when applied to a specific information provider such as an encyclopedia, and even when applied to various information providers. A user interface for providing user-friendly information can be realized.

[Modification]
In each of the above-described embodiments and examples, an example in which the dialogue server 1 acquires information data from the information providing site 5 connected via the communication network 3 is shown. However, the information data may be acquired from a database held in the dialogue server 1. In this case, the data acquisition unit 31 may determine the information request destination from the information providing site 5 and the database in the dialogue server 1.

  Further, in each of the above-described embodiments and examples, the example in which the utterance text data acquired by the utterance data acquisition unit 35 is generated from the voice uttered by the user is shown. However, the comment text data may be text data input by a user operation via the input device of the client 2 such as a keyboard or a touch panel. Further, the output form of the information providing message data and the dialogue message data is not limited to the voice output, but may be displayed on a display device included in the client 2 or may be displayed on a printing device connected to the client 2. It may be printed. Further, the information providing message data and the dialogue message data may be output in two or more output forms. For example, the information providing message data may be displayed on a display device while being output as voice.

  Further, the data output unit 34 described above may transmit information data extracted from the information content indicated by the information providing message to the client 2 together with the information providing message data. In this case, the user can visually recognize the entire information data extracted from the information providing site 5 or the database via the screen, and can also hear a voice reading the information content included in the information data.

  In the plurality of flowcharts used in the above description, a plurality of steps (processes) are described in order, but the execution order of the steps executed in the present embodiment is not limited to the description order. In the present embodiment, the order of the illustrated steps can be changed within a range that does not hinder the contents. Moreover, each above-mentioned embodiment, each Example, and each modification can be combined in the range in which the content does not conflict.

  A part or all of each of the above embodiments, examples, and modifications may be specified as in the following supplementary notes. However, each embodiment, each Example, and each modification is not limited to the following description.

(Appendix 1)
A data acquisition unit that acquires information data extracted according to at least one search keyword;
An extraction unit for extracting a related keyword related to the at least one search keyword from the information data acquired by the data acquisition unit;
A message generation unit that generates dialog message data including related keywords extracted by the extraction unit;
A data output unit for outputting dialogue message data generated by the message generation unit;
An information providing apparatus comprising:

(Appendix 2)
A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The data acquisition unit acquires new information data related to the related keyword included in the dialogue message data according to the response content indicated by the response message data acquired by the response message data acquisition unit.
The information providing apparatus according to appendix 1.

(Appendix 3)
The extraction unit further extracts text data indicating information content matching the at least one search keyword included in the information data acquired by the data acquisition unit;
The message generator further generates information providing message data including the text data extracted by the extractor;
The data output unit further outputs the dialogue message data after outputting the information providing message.
The information providing apparatus according to appendix 1 or 2.

(Appendix 4)
A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The extraction unit is configured to obtain information about the related keyword included in the dialogue message data from information data acquired by the data acquisition unit according to a response content indicated by the response message data acquired by the response message data acquisition unit. Extract further text data indicating information content,
The message generator further generates information providing message data including the text data extracted by the extractor;
The data output unit further outputs the information providing message generated for the related keyword.
The information providing apparatus according to any one of supplementary notes 1 to 3.

(Appendix 5)
The message generation unit includes the information provision message data including partial text data in which the text data extracted by the extraction unit is separated at a predetermined position, and a confirmation for confirming whether it is necessary to continue the information provision Generate message data,
The data output unit outputs the information providing message data including the partial text data and the confirmation message data;
The information providing apparatus according to appendix 3 or 4.

(Appendix 6)
A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The extraction unit extracts a plurality of related keywords related to the at least one search keyword from the information data acquired by the data acquisition unit,
The message generation unit holds information that makes it possible to identify a related keyword that has been selected for inclusion in the dialogue message data and other related keywords from among the plurality of related keywords, and the held information And an unselected related keyword is selected from the plurality of related keywords based on the response content indicated by the response message data acquired by the response message data acquisition unit, and the selected new related keyword Generate new interactive message data including
The data output unit outputs the new dialogue message data generated by the message generation unit after receiving the response message data by the response message data acquisition unit.
The information providing apparatus according to any one of appendices 1 to 5.

(Appendix 7)
A request message data acquisition unit for acquiring request message data indicating an information request from a user including at least one search keyword;
Further comprising
The data acquisition unit acquires information data extracted by the at least one search keyword included in the request message data according to the request message data acquired by the request message data acquisition unit,
The extraction unit extracts a plurality of related keywords related to the at least one search keyword from the information data acquired by the data acquisition unit,
The message generation unit holds information that makes it possible to identify a related keyword selected for inclusion in the dialogue message data and other related keywords from among the plurality of related keywords, and obtains the request message data When the request utterance data after the second time indicating the same request content from the same user is acquired by the section, the dialogue message data including the unselected related keyword is generated based on the held information.
The information providing apparatus according to any one of appendices 1 to 6.

(Appendix 8)
At least one computer
Obtain information data extracted according to at least one search keyword,
Extracting a related keyword related to the at least one search keyword from the acquired information data;
Generating interactive message data including the extracted related keywords;
Outputting the generated dialogue message data;
Information providing method including

(Appendix 9)
The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
According to the response content indicated by the response message data, new information data relating to the related keyword included in the dialogue message data is acquired.
The information providing method according to appendix 8, further including:

(Appendix 10)
The at least one computer comprises:
Extracting text data indicating information content that matches the at least one search keyword included in the acquired information data;
Generating information providing message data including the extracted text data;
Outputting the dialogue message data after outputting the information providing message;
The information provision method according to appendix 8 or 9, further including:

(Appendix 11)
The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
In accordance with the response content indicated by the response message data, text data indicating the information content about the related keyword included in the dialogue message data is extracted from the acquired information data,
Generating information providing message data including the extracted text data;
Outputting the information providing message generated for the related keyword;
The information providing method according to any one of appendices 8 to 10, further including:

(Appendix 12)
The at least one computer comprises:
Generating information providing message data including partial text data in which the extracted text data is separated at a predetermined position, and confirmation message data for confirming the necessity of continuing information provision;
Outputting the information providing message data including the partial text data and the confirmation message data;
The information provision method according to appendix 10 or 11, further including:

(Appendix 13)
The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
Based on the held information and the response content indicated by the response message data, an unselected related keyword is selected from the plurality of related keywords,
Generating new dialogue message data including the selected new related keyword;
After the response message data is received, the new dialog message data generated by the message generation unit is output.
The information provision method according to any one of appendices 8 to 12, further including:

(Appendix 14)
The at least one computer comprises:
Obtaining request utterance data indicating an information request from a user including at least one search keyword;
In response to the request message data, obtain information data extracted by the at least one search keyword included in the request message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
When the second and subsequent request message data indicating the same request content from the same user is acquired, the dialog message data including unselected related keywords is generated based on the held information.
The information provision method according to any one of appendices 8 to 13, further including:

(Appendix 15)
On at least one computer,
Obtain information data extracted according to at least one search keyword,
Extracting a related keyword related to the at least one search keyword from the acquired information data;
Generating interactive message data including the extracted related keywords;
Outputting the generated dialogue message data;
A program that lets you do that.

(Appendix 16)
Said at least one computer,
Receiving response message data returned in response to the dialogue message data;
According to the response content indicated by the response message data, new information data relating to the related keyword included in the dialogue message data is acquired.
The program according to supplementary note 15, further causing the above to be executed.

(Appendix 17)
Said at least one computer,
Extracting text data indicating information content that matches the at least one search keyword included in the acquired information data;
Generating information providing message data including the extracted text data;
Outputting the dialogue message data after outputting the information providing message;
The program according to appendix 15 or 16, further causing the above to be executed.

(Appendix 18)
Said at least one computer,
Receiving response message data returned in response to the dialogue message data;
In accordance with the response content indicated by the response message data, text data indicating the information content about the related keyword included in the dialogue message data is extracted from the acquired information data,
Generating information providing message data including the extracted text data;
Outputting the information providing message generated for the related keyword;
18. The program according to any one of appendices 15 to 17, further executing the above.

(Appendix 19)
Said at least one computer,
Generating information providing message data including partial text data in which the extracted text data is separated at a predetermined position, and confirmation message data for confirming the necessity of continuing information provision;
Outputting the information providing message data including the partial text data and the confirmation message data;
The program according to appendix 17 or 18, further executing the above.

(Appendix 20)
Said at least one computer,
Receiving response message data returned in response to the dialogue message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
Based on the held information and the response content indicated by the response message data, an unselected related keyword is selected from the plurality of related keywords,
Generating new dialogue message data including the selected new related keyword;
After the response message data is received, the new dialog message data generated by the message generation unit is output.
The program according to any one of supplementary notes 15 to 19, further executing the above.

(Appendix 21)
Said at least one computer,
Obtaining request utterance data indicating an information request from a user including at least one search keyword;
In response to the request message data, obtain information data extracted by the at least one search keyword included in the request message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
When the second and subsequent request message data indicating the same request content from the same user is acquired, the dialog message data including unselected related keywords is generated based on the held information.
The program according to any one of supplementary notes 15 to 20, further causing the above to be executed.

1 Voice dialogue server device (dialogue server)
2 Client device (client)
5 Information service site 10 CPU
DESCRIPTION OF SYMBOLS 11 Memory 17 Voice input part 18 Voice output part 21 Voice recognition part 22 Speech synthesizer 25 Dialog control engine 26 WEB server processing part 31 Data acquisition part 32 Extraction part 33 Message generation part 34 Data output part 35 Speech data acquisition part 61 Request holding Part 100 information providing device

Claims (15)

A data acquisition unit that acquires information data extracted according to at least one search keyword;
An extraction unit for extracting a related keyword related to the at least one search keyword from the information data acquired by the data acquisition unit;
A message generation unit that generates dialog message data including related keywords extracted by the extraction unit;
A data output unit for outputting dialogue message data generated by the message generation unit;
An information providing apparatus comprising: A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The data acquisition unit acquires new information data related to the related keyword included in the dialogue message data according to the response content indicated by the response message data acquired by the response message data acquisition unit.
The information providing apparatus according to claim 1. The extraction unit further extracts text data indicating information content matching the at least one search keyword included in the information data acquired by the data acquisition unit;
The message generator further generates information providing message data including the text data extracted by the extractor;
The data output unit further outputs the dialogue message data after outputting the information providing message.
The information providing apparatus according to claim 1 or 2. A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The extraction unit is configured to obtain information about the related keyword included in the dialogue message data from information data acquired by the data acquisition unit according to a response content indicated by the response message data acquired by the response message data acquisition unit. Extract further text data indicating information content,
The message generator further generates information providing message data including the text data extracted by the extractor;
The data output unit further outputs the information providing message generated for the related keyword.
The information providing apparatus according to any one of claims 1 to 3. The message generation unit includes the information provision message data including partial text data in which the text data extracted by the extraction unit is separated at a predetermined position, and a confirmation for confirming whether it is necessary to continue the information provision Generate message data,
The data output unit outputs the information providing message data including the partial text data and the confirmation message data;
The information providing apparatus according to claim 3 or 4. A response message data acquisition unit for receiving response message data returned in response to the dialog message data;
Further comprising
The extraction unit extracts a plurality of related keywords related to the at least one search keyword from the information data acquired by the data acquisition unit,
The message generation unit holds information that makes it possible to identify a related keyword that has been selected for inclusion in the dialogue message data and other related keywords from among the plurality of related keywords, and the held information And an unselected related keyword is selected from the plurality of related keywords based on the response content indicated by the response message data acquired by the response message data acquisition unit, and the selected new related keyword Generate new interactive message data including
The data output unit outputs the new dialogue message data generated by the message generation unit after receiving the response message data by the response message data acquisition unit.
The information providing apparatus according to any one of claims 1 to 5. A request message data acquisition unit for acquiring request message data indicating an information request from a user including at least one search keyword;
Further comprising
The data acquisition unit acquires information data extracted by the at least one search keyword included in the request message data according to the request message data acquired by the request message data acquisition unit,
The extraction unit extracts a plurality of related keywords related to the at least one search keyword from the information data acquired by the data acquisition unit,
The message generation unit holds information that makes it possible to identify a related keyword selected for inclusion in the dialogue message data and other related keywords from among the plurality of related keywords, and obtains the request message data When the request utterance data after the second time indicating the same request content from the same user is acquired by the section, the dialogue message data including the unselected related keyword is generated based on the held information.
The information providing apparatus according to any one of claims 1 to 6. At least one computer
Obtain information data extracted according to at least one search keyword,
Extracting a related keyword related to the at least one search keyword from the acquired information data;
Generating interactive message data including the extracted related keywords;
Outputting the generated dialogue message data;
Information providing method including The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
According to the response content indicated by the response message data, new information data relating to the related keyword included in the dialogue message data is acquired.
The information providing method according to claim 8, further comprising: The at least one computer comprises:
Extracting text data indicating information content that matches the at least one search keyword included in the acquired information data;
Generating information providing message data including the extracted text data;
Outputting the dialogue message data after outputting the information providing message;
The information providing method according to claim 8 or 9, further comprising: The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
In accordance with the response content indicated by the response message data, text data indicating the information content about the related keyword included in the dialogue message data is extracted from the acquired information data,
Generating information providing message data including the extracted text data;
Outputting the information providing message generated for the related keyword;
The information providing method according to any one of claims 8 to 10, further comprising: The at least one computer comprises:
Generating information providing message data including partial text data in which the extracted text data is separated at a predetermined position, and confirmation message data for confirming the necessity of continuing information provision;
Outputting the information providing message data including the partial text data and the confirmation message data;
The information providing method according to claim 10 or 11, further comprising: The at least one computer comprises:
Receiving response message data returned in response to the dialogue message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
Based on the held information and the response content indicated by the response message data, an unselected related keyword is selected from the plurality of related keywords,
Generating new dialogue message data including the selected new related keyword;
After the response message data is received, the new dialog message data generated by the message generation unit is output.
The information provision method according to any one of claims 8 to 12, further comprising: The at least one computer comprises:
Obtaining request utterance data indicating an information request from a user including at least one search keyword;
In response to the request message data, obtain information data extracted by the at least one search keyword included in the request message data;
Extracting a plurality of related keywords related to the at least one search keyword from the acquired information data;
Holding information that makes it possible to identify a related keyword selected for inclusion in the interactive message data and other related keywords among the plurality of related keywords;
When the second and subsequent request message data indicating the same request content from the same user is acquired, the dialog message data including unselected related keywords is generated based on the held information.
The information provision method according to any one of claims 8 to 13, further comprising: On at least one computer,
Obtain information data extracted according to at least one search keyword,
Extracting a related keyword related to the at least one search keyword from the acquired information data;
Generating interactive message data including the extracted related keywords;
Outputting the generated dialogue message data;
A program that lets you do that.

Images