JP2017134761A - Information processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device capable of making a response that is acquired by a user in keyword retrieval.SOLUTION: The information processing device for retrieving an inputted keyword stores information to the keyword, retrieves the information to the keyword, determines whether the information to the keyword is an unknown word or a known word on the basis of a retrieval result due to the retrieval, performs web retrieval of the keyword to retrieve a category with the keyword belonging thereto when the keyword is determined to be the unknown word by the determination, and associates the keyword with the category for registration.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information processing apparatus.

  Conventionally, first, voice information is converted into text information, and the text information is searched. Next, if the amount of information retrieved from the text information is greater than or equal to the threshold value, audio information for narrowing down the amount of information is requested, and a predetermined amount of information is received from the server that holds the information corresponding to the attribute obtained from the text information. There is known a method of searching for (for example, Patent Document 1). When such a method is used, a search can be continued following an incorrect answer.

JP2015-69102A

  However, in the above-described conventional technology, the keyword to be searched may not be in the search range. In this case, there are many cases where the response requested by the user cannot be obtained as a search result.

  In view of the above problems, an object of the present invention is to provide an information processing apparatus that can respond to a user's request in keyword search.

In order to achieve the above object, in one embodiment, an information processing apparatus for searching for an input keyword is:
A storage unit for storing information on the keyword;
A first search unit for searching the information for the keyword;
Judging whether the keyword is an unknown word that does not have information on the keyword in the storage unit or whether the information on the keyword is a known word in the storage unit based on a search result by the first search unit And
When the determination unit determines that the keyword is the unknown word, a second search unit that performs a Web search for the keyword and searches for a category to which the keyword belongs;
A registration unit that registers the keyword and the category in the storage unit in association with each other.

  According to this aspect, the storage unit can store information on the input keyword. And the said 1st search part can search the information which this said memory | storage part memorize | stores. Therefore, when information on a keyword is stored in the storage unit, that is, when the determination unit determines that the keyword is a known word, the information processing apparatus can make a response to the keyword.

  On the other hand, when a keyword having no information is input to the storage unit, that is, when the determination unit determines that the keyword is an unknown word, the second search unit searches the keyword for the Web, The category to which the keyword belongs is searched. Since the second search unit is a Web search, the category to which the keyword belongs may be searched even if there is no information in the storage unit. In this way, when the category to which the keyword belongs is found by Web search, the category to which the keyword belongs can be specified. Therefore, the registration unit can register the keyword and the category in the storage unit in association with each other. When the keyword and the category are registered in association with each other, even if the keyword is an unknown word, when the information stored in the storage unit is searched from the next time, there is information on the keyword. That is, the information processing apparatus can handle the keyword in the same way as a known word. Therefore, after the registration, when the first search unit searches for information stored in the storage unit, a search result similar to a known word can be displayed. Therefore, even if an unknown word is input, the information processing apparatus can display the same search result as the known word and can respond to the user's request.

  According to the present embodiment, it is possible to provide an information processing apparatus that can respond to a user's request in keyword search.

1 is an overall configuration diagram illustrating an example of a system configuration using an information processing apparatus according to an embodiment of the present invention. It is a flowchart which shows an example of the process performed by the information processing apparatus which concerns on one Embodiment of this invention. It is a functional block diagram which shows an example of a function structure of the information processing apparatus which concerns on one Embodiment of this invention.

  Hereinafter, embodiments for carrying out the invention will be described with reference to the drawings.

  FIG. 1 is an overall configuration diagram showing an example of a system configuration using an information processing apparatus according to an embodiment of the present invention. In the illustrated example, the user UR inputs the keyword KEY and the like to the agent system 10 which is an example of the information processing apparatus, and the agent system 10 outputs the search result.

  The information processing apparatus has an input device such as a microphone or a keyboard for inputting words including the keyword KEY. In addition, the information processing apparatus includes an arithmetic device that performs calculation and control for realizing processing for searching for the keyword KEY, a CPU (Central Processing Unit) that is a control device, and the like. Furthermore, the information processing apparatus includes a network I / F (interface) or the like for connecting to the external system 11 via a network. Furthermore, the information processing apparatus includes a storage device that stores data indicating information on the keyword.

  In the illustrated configuration, in step S1, the agent system 10 performs a first reception process for inputting the keyword KEY. On the other hand, the agent system 10 has a DB (database) that stores information corresponding to the input keyword KEY.

  Next, in step S2, the agent system 10 performs a first search process for searching for information stored in the DB for the keyword KEY input in step S1. Subsequently, if there is information corresponding to the DB as a result of the search, the agent system 10 outputs information corresponding to the user UR. On the other hand, if there is no corresponding information, the agent system 10 uses the external system 11 to specify the category to which the keyword KEY belongs by performing a Web search. In the illustrated example, the specified category is an example of a POI (point of interest) category.

  In step S3, the agent system 10 performs a first response process for displaying the search result of the first search process.

  Hereinafter, a process performed when the first search process results in a search result indicating that there is no corresponding information will be described. In the following description, a keyword for which there is no relevant information by the first search process, that is, there is no information in the DB of the agent system 10 is referred to as an “unknown word”, while there is relevant information, A keyword having information in the DB of the agent system 10 is referred to as “known word”.

  FIG. 2 is a flowchart illustrating an example of processing performed by the information processing apparatus according to the embodiment of the present invention.

  In step S10, the agent system accepts a keyword or the like by the first acceptance process (FIG. 1). For example, in the example shown in FIG. 1, a word including a keyword is input by voice uttered by the user. In this example, the agent system inputs voice with a microphone or the like and converts it into an electrical signal. Then, a keyword is extracted from words including the keyword by voice recognition or the like from the voice uttered by the user. Thereafter, each process is performed on the extracted keyword.

  In step S11, the agent system searches the keyword input in step S10 by the first search unit or the like. The search range is a range of information stored in the known word keyword DB that is an example of the storage unit. That is, in step S11, the agent system compares the input keyword with each information stored in the known word keyword DB. If the information stored in the known word keyword DB includes information about the keyword, the search result is “there is corresponding information”. On the other hand, if the information stored in the known word keyword DB does not include information related to the keyword, the search result is “no corresponding information”. In this way, a keyword whose search result is “no corresponding information” is an unknown word.

  In step S12, the agent system determines whether there is corresponding information based on the search result by the determination unit or the like. That is, in step S12, the agent system determines whether there is information for the keyword in the information stored in the known word keyword DB based on the search result in step S11. Next, when there is corresponding information (YES in step S12), the agent system proceeds to step S13. On the other hand, if there is no corresponding information (NO in step S12), the agent system proceeds to step S14.

  In step S13, the agent system displays the search result. For example, the agent system displays a keyword in the known word keyword DB. In addition, the agent system may display information corresponding to a keyword in the known word keyword DB or information as a candidate.

  In step S14, the agent system stores the keyword in the unknown word keyword DB. For example, in the example shown in FIG. 1, the word “I want to buy a diaper” is input. Of these, “diaper” is a keyword, and “diaper” is an unknown word. In this case, “diaper” is stored in the unknown word keyword DB. As shown in FIG. 1, as in the first response process in step S3, the agent system displays “Not found” or the like to notify the user that the keyword is an unknown word. Also good.

  In step S15, the agent system searches for an unknown word using the second search unit or the like. Specifically, in step S15, the agent system performs a Web search using the external system 11 (FIG. 1) such as Wikipedia (registered trademark). Note that step S15 is performed by, for example, batch activation. That is, in step S15, the agent system searches for a range different from the range searched in step S11. Specifically, in step S15, a dictionary DB such as Wikipedia (registered trademark) is preferably used for the search. For example, since the known word keyword DB is realized by a storage device or the like included in the agent system, in step S11, data stored in the storage device or the like is a search target. On the other hand, in step S15, a DB or the like realized by the external system 11 is used for the search. That is, in step S15, data on the Web is a search target via the network.

  In step S16, the agent system acquires the category to which the unknown word belongs from the search result in step S15. Specifically, the category acquired from the result of searching for “diapers” in step S15 indicates, for example, a group in which other items used in a scene similar to the scene where “diapers” are used are collected. “Childcare items”, “welfare items” or “daily items”. That is, the category is a word indicating the classification to which the keyword belongs. The specified category is preferably the POI category. In the case of the POI category, it is possible to accurately search for stores or the like that handle items to be searched for searching the map data.

  In step S17, the agent system compares the known word in the known word keyword DB with the category acquired in step S16.

  In step S18, the agent system determines whether the category information is in the known word keyword DB as a result of the comparison in step S17. That is, in step S17, when any known word in the known word keyword DB matches the category, the agent system determines that there is category information. Next, when there is category information (YES in step S18), the agent system proceeds to step S19. On the other hand, if there is no category information (NO in step S18), the agent system proceeds to step S20.

  In step S19, the agent system registers keywords and the like. That is, in step S19, the agent system associates the category acquired in step S16 with the keyword that is an unknown word and registers it in the known word keyword DB.

  Specifically, it is assumed that “daily necessities” data is stored in advance in the known word keyword DB. In this case, in step S17, the category of “daily necessities” acquired in step S16 matches the known word “daily necessities” indicated by the data in the known word keyword DB. In this way, when the result matches the category and the known word, “diapers” and “daily necessities” are linked and registered in the known word keyword DB by the agent system.

  When the above processing is performed, a keyword that is an unknown word can be searched in the same manner as a known word. Specifically, as shown in FIG. 1, after the process shown in FIG. 2 is performed, for example, when the second reception process in step S4 is performed by the agent system in the same manner as the first reception process in step S1. The agent system performs the second search process in step S5. In the second search process, since the keyword that was an unknown word is registered when the first search process is performed, the keyword is processed in the same manner as the known word. Therefore, in the second search process, a keyword that was an unknown word can be searched as a known word. Therefore, the agent system can display a search similar to the result of searching for a known word in the second response in step S6.

  In step S20, the agent system associates the category acquired in step S16 with the keyword that is an unknown word, and newly creates new category information in the known word keyword DB. In this way, the agent system registers the category and the keyword that is an unknown word in association with the known word keyword DB.

  After the registration, as shown in FIG. 1, in step S4, as in the first reception process, when a second reception process is entered in which "I want to buy a diaper" is performed, in step S5, the agent system The “daily necessities” can be searched by the second search process. Therefore, the agent system can acquire POI data of a daily necessities store where “daily necessities” are sold. As a result, in step S6, the agent system can make a response requested by the user such as “searched for nearby daily necessities” by the second response process. In step S6, based on the POI data, a map or guidance showing the location of the daily necessities store may be displayed by the agent system.

  FIG. 3 is a functional block diagram illustrating an example of a functional configuration of the information processing apparatus according to the embodiment of the present invention. As illustrated, the information processing apparatus includes a storage unit FM, a first search unit F1S, a determination unit FJ, a second search unit F2S such as a POI category specifying unit, and a registration unit FR.

  The storage unit FM stores information on the input keyword KEY. Note that the storage unit FM is realized by a storage device or the like included in the information processing apparatus.

  The first search unit F1S searches for information on the keyword KEY that is input from information stored in the storage unit FM. The first search unit F1S is realized by a CPU, a storage device, and the like included in the information processing apparatus.

  The determination unit FJ determines whether the keyword KEY is an unknown word or a known word based on the search result by the first search unit F1S. Note that the determination unit FJ is realized by a CPU or the like included in the information processing apparatus.

  When the determination unit FJ determines that the keyword KEY is an unknown word, the second search unit F2S uses the external system 11 or the like to search the Web for the input keyword KEY. By this Web search, the second search unit F2S searches the category to which the keyword KEY belongs. Note that the second search unit F2S is realized by a CPU, a storage device, and the like included in the information processing apparatus.

  The registration unit FR registers the keyword KEY and the category specified by the search by the second search unit F2S in association with each other. The registration unit FR is realized by a CPU, a storage device, and the like included in the information processing apparatus.

  Thus, the registration unit FR registers the category specified by the search by the second search unit F2S and the keyword KEY of the unknown word in association with each other. Therefore, even if an unknown word keyword KEY is input, the keyword will be treated as a known word in the next search. Therefore, even if a keyword that is an unknown word is searched, a category is associated with the keyword, so that the information processing apparatus can search in the same manner as a known word. That is, the information processing apparatus can make a response requested by the user, such as displaying information on a store that handles the item indicated by the keyword, in searching for the keyword.

  As mentioned above, although the form for implementing this invention was explained in full detail, this invention is not limited to this specific embodiment, In the range of the summary of this invention described in the claim, various It is possible to modify or change.

10. Agent system (information processing device example)
11 External System KEY Keyword FM Storage Unit F1S First Search Unit F2S Second Search Unit FJ Judgment Unit FR Registration Unit

Claims (1)

An information processing apparatus for searching for an input keyword,
A storage unit for storing information on the keyword;
A first search unit for searching the information for the keyword;
Judging whether the keyword is an unknown word that does not have information on the keyword in the storage unit or whether the information on the keyword is a known word in the storage unit based on a search result by the first search unit And
When the determination unit determines that the keyword is the unknown word, a second search unit that performs a Web search for the keyword and searches for a category to which the keyword belongs;
An information processing apparatus including a registration unit that links the keyword and the category and registers the keyword and the category in the storage unit.

Images