JP6993913B2 - Information processing equipment, information processing methods, and programs - Google Patents

Description

The present invention relates to an information processing apparatus, an information processing method, and a program.

Traditionally, a query querying a given fact entity is received from a client device, the electronic resource associated with the received query is identified, and knowledge of the given fact entity from the electronic resource associated with the given fact entity. A method of selecting content to be displayed in a panel is disclosed (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2016-139438

In the conventional technique, when multiple entities are related to the input query, it may not be possible to select the appropriate entity. For this reason, it is desirable to properly evaluate the relationship between the entity and the keyword so that the appropriate entity is selected based on the entered query.

The present invention has taken such circumstances into consideration, and one of the objects of the present invention is to provide an information processing device, an information processing method, and a program capable of appropriately evaluating the relationship between an entity and a keyword. do.

One aspect of the present invention is a determination unit that refers to the attribute information associated with the entity and determines whether the entity is the application target of the first criterion or the entity is the application target of the second criterion. , When it is determined that the first criterion is applied, the combination of the entity and the keyword is evaluated based on the first criterion, and it is determined that the second criterion is applied. , An information processing apparatus including an evaluation unit that evaluates the combination based on the second criterion.

According to one aspect of the present invention, the relationship between the entity and the keyword can be appropriately evaluated. As a result, the appropriate entity can be selected and provide the user with more appropriate information.

It is a figure which shows an example of the usage environment in which a knowledge data server is used. It is a figure which shows a part of the knowledge graph 72 schematically. It is a figure which shows an example of the content which can be referred by tracing the location information associated with the entity "baseball player A". It is a flowchart which shows an example of the flow of the process executed by the knowledge data server 50. It is a figure which shows an example of the contents included in the 2nd score information 76 before the score is given. It is a figure which shows an example of the plurality of location information associated with the entity to which the 2nd standard is applied. It is a figure which shows an example of the content of the information of the number of accesses to a plurality of specific contents. It is a figure which shows an example of the content of the 2nd score information 76 to which the score was given. It is a flowchart which shows an example of the flow of the process which provides a knowledge panel to a user. It is a conceptual diagram of the process flow which provides a knowledge panel to a user. It is a figure which shows an example of the knowledge panel provided to a user. It is a figure which shows an example of the content of the 2nd score information 76A to which auxiliary information is attached. It is a figure which shows an example of the content of the reference information 78.

Hereinafter, embodiments of the information processing apparatus, information processing method, and program of the present invention will be described with reference to the drawings.

The information processing device is realized by one or more processors. An information processing device is a device that evaluates a combination of an entity and a keyword. An entity represents an entity or concept of an object and is a part of a knowledge graph that describes the correspondence between the entities. The knowledge graph is associated with the attribute information of the entity. For example, when a query is entered, if the query corresponds to an entity, it is possible to return a wealth of information to the user rather than a simple keyword search. However, since there are different entities in the Knowledge Graph that contain the same keyword, it may be difficult to immediately identify the relevant entity from the keywords entered as a query.

The information processing device refers to the attribute information associated with the entity, determines whether the entity is the application target of the first standard or the entity is the application target of the second standard, and applies the first standard. If it is determined that the combination is based on the first criterion, the combination of the entity and the keyword is evaluated, and if it is determined that the second criterion is applied, the above combination is based on the second criterion. To evaluate. The second criterion is a criterion that can more appropriately evaluate the entity determined to be the target of the application of the second criterion as compared with the first criterion. As a result, the information processing apparatus can appropriately evaluate the relationship between the entity and the keyword.

For example, an entity to be evaluated based on the second criterion is, for example, an entity in which a concept that is the same as, similar to, or includes a part of the entity is commonly used in a plurality of media (mediums). be. For example, entities related to entertainment (eg, comics, anime, movies, television (eg, TV dramas), novels, music, etc.) are subject to evaluation based on the second criterion. The media is, for example, comics, anime, movies, television dramas, novels, music, and the like. Specifically, for example, if the original is a manga called "OO Man" and this manga is developed as an anime, drama, movie, or novel, "OO Man Manga" or "OO Man" Entities such as "Man Anime" ... "○○ Man Novel" and entities that summarize "○○ Man Manga" to "○○ Man Novel" are subject to evaluation based on the second criterion.

<First Embodiment>
FIG. 1 is a diagram showing an example of a usage environment in which a knowledge data server is used. The usage environment of the knowledge data server includes, for example, one or more terminal devices 10, one or more search servers 20, a crawl target device 30, and a knowledge data server 50. The search server 20 and the knowledge data server 50 may be configured as one.

These devices and the like communicate with each other via the network NW. The network NW includes, for example, a WAN (Wide Area Network), a LAN (Local Area Network), the Internet, a dedicated line, a wireless base station, a provider, and the like.

[Terminal device]
The terminal device 10 is, for example, a device used by a user, for example, a mobile phone such as a smartphone, a tablet computer, a notebook computer, a desktop computer, or the like. The user operates the terminal device 10, inputs a query, and sends the input query to the search server 20.

[Search server]
The search server 20 acquires a query from the terminal device 10 and provides the terminal device 10 with information corresponding to the acquired query. The search server 20 searches for information related to the acquired query by using a search engine. Further, the search server 20 acquires information related to the acquired query from the knowledge data server 50. The search server 20 provides the search result of the search engine and the information acquired from the knowledge data server 50 to the terminal device 10.

[Crawling target device]
The crawl target device 30 is a device that provides information to the knowledge data server 50 in response to a request from the knowledge data server 50.

[Knowledge Data Server]
The knowledge data server 50 generates a knowledge graph 72 based on data provided by, for example, a crawl target device 30 or the like.

The knowledge data server 50 includes, for example, a communication unit 52, a collection unit 54, a determination unit 56, an evaluation unit 58, a response unit 60, and a storage unit 70. The determination unit 56, the collection unit 54, the evaluation unit 58, and the response unit 60 are realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program stored in the storage device. Further, these functional units may be realized by hardware such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), etc. It may be realized by the collaboration of software and hardware. Further, the above program may be stored in a storage device in advance, or is stored in a removable storage medium such as a DVD or a CD-ROM, and the storage medium is attached to the drive device of the knowledge data server 50. It may be installed in the storage device.

The storage unit 70 is realized by, for example, a ROM (Read Only Memory), a flash memory, an SD card, a RAM (Random Access Memory), an HDD (Hard Disc Drive), a register, or the like. Further, a part or all of the storage unit 70 may be NAS (Network Attached Storage), an external storage server device, or the like. For example, the knowledge graph 72, the first score information 74, and the second score information 76 are stored in the storage unit 70. Details of this information will be described later.

The knowledge graph 72 is information associated with the location information of the entity, the class, the property, and the content related to the entity. FIG. 2 is a diagram schematically showing a part of the Knowledge Graph 72.

As shown in FIG. 2, in an entity, information that can identify each entity (hereinafter referred to as entity identification information), an entity name, location information of contents related to the entity, and a class are associated with each other. Has been done. The location information is information for specifying a position on the Web, and is, for example, a URL (Uniform Resource Locator). The entity E1 of the knowledge graph 72 includes the entity identification information EID "0001", the entity name "baseball player A", and the location information of the content related to the entity "http: // encyclopedia content / baseball player". The URL "A" is associated with the class "01". In the example of FIG. 2, the entity E1 and the entity E2 belong to the class with the property name "affiliated team".

An entity may represent, for example, an entity of an object (eg, an object that exists in the real world), or a concept of an object (eg, a concept defined in the real world or virtual world). It may be represented. For example, when the object is the concept of "building", the entity may represent an entity such as "○○ tower" or "○○ building". Further, for example, when the object is the concept of "economics", the entity may represent an insubstantial abstract concept such as "microeconomics" or "macroeconomics".

The things described in Knowledge Graph 72 are defined by the ontology. An ontology is a definition of a class and a property of an object, and is a collection of constraints that hold between the class and the property.

A class is a group of things that have the same properties in an ontology. What the nature of an object is, that is, which class the object belongs to, is determined by the properties described below. A class is an example of attribute information of an entity.

For example, things that have a beak, an oviparous vertebrate, and forelimbs that are wings are classified in the "bird" class. Also, in the class of "birds", things that have the property of not being able to fly are classified into lower classes such as "penguins" and "ostriches". In this way, the class system may have a hierarchical structure having a higher-lower relationship. The nature of the upper class is inherited by the lower class. In the example above, the "bird" class's "beak-bearing, oviparous vertebrate with winged forelimbs" property is a lower class property of "penguins" and "ostriches". Will also be included. For example, if the class name for identifying the class is a class "bird", the class name may be represented by the character string "bird". The class name does not necessarily have to represent a meaning, and for example, even in the class "bird", a character string indicating mere identification information such as "information 1" or "C1" may be assigned. .. The above-mentioned entity, that is, an entity, belongs to one of the classes included in the class system defined by the ontology.

Properties are attributes that describe the nature and characteristics of things and the relationships between classes. For example, a property may be an attribute that has the property of "having ... as a component of the body" or the property of "living in", or "a class is a higher class and a certain class is a lower class". It may be an attribute indicating the relationship between the upper and lower levels of "class". As for the property name for identifying the property, the property name itself may or may not represent the meaning, as in the class name described above.

The Knowledge Graph 72 is a directed graph in which the above-mentioned classes are represented as nodes and the above-mentioned properties are represented as labeled and directional edges. With such a graph structure, information about things can be discriminated by nodes, and relationships between things can be discriminated by edges.

FIG. 3 is a diagram showing an example of content that can be referred to by tracing the location information associated with the entity “baseball player A”. The content of this location information is information provided by a server device (not shown) that provides a service of an encyclopedia that can be edited by a user. This location information indicates the location of the item of "baseball player A" including the information of "baseball player A" in the above encyclopedia.

Returning to the description of FIG. The communication unit 52 includes a communication interface such as a network interface card. The collecting unit 54 is realized by a so-called crawler program. The information collected by the collecting unit 54 is, for example, data that is on the network NW (in the storage area of the crawling target device 30) and can be viewed by a browser. The collecting unit 54 collects the content for constructing the Knowledge Graph 72. Further, the collecting unit 54 acquires information on the user's usage status for services in the network NW such as the user's search log. In addition, in the construction of the knowledge graph 72, the data provided by the predetermined server may be used instead of (or in addition to) the content collected by the collecting unit 54. For example, the knowledge data server 50 may download the data used for constructing the knowledge graph 72 from the data provider and construct the knowledge graph 72 using the downloaded data.

Hereinafter, the preparatory process for responding to the inquiry from the search server 20 will be described. The search server 20 may transfer the query input from the terminal device 10 to the knowledge data server 50 and inquire whether or not there is a corresponding entity. The knowledge data server 50 performs the following preparatory processing in order to respond to this.

For example, the determination unit 56 refers to the class associated with the entity for all the entities registered in the Knowledge Graph 72, and whether the entity is the application target of the first criterion or the application target of the second criterion. Determine if there is.

When the determination unit 56 determines that the entity is the target of application of the first criterion, the evaluation unit 58 evaluates the combination of the entity and the keyword based on the first criterion, and the determination unit 56 evaluates the entity. If it is determined that the two criteria are applicable, the combination of the entity and the keyword is evaluated based on the second criterion. The keyword is, for example, extracted from the search log of the search server 20 or extracted from the content collected by the collecting unit 54. Hereinafter, the keyword to be evaluated by the evaluation unit 58 will be referred to as a target keyword.

[flowchart]
FIG. 4 is a flowchart showing an example of the flow of processing executed by the knowledge data server 50. This flowchart is executed, for example, at predetermined intervals. First, the determination unit 56 selects one entity to be processed (S10). The entity to be processed is an entity included in Knowledge Graph 72. Next, the determination unit 56 refers to the class of the selected entity (S12). Next, the determination unit 56 determines whether or not the class is a specific class (S14). The specific class is a class indicating an entity to be evaluated based on the second criterion.

If the class is not a specific class, the evaluation unit 58 executes a process based on the first criterion (S16), and if the class is a specific class, executes a process based on the second criterion (S18). The processing based on the first standard and the processing based on the second standard will be described later.

Next, the determination unit 56 determines whether or not all the entities to be processed have been selected (S20). If all the entities are not selected, the process returns to step S10. When all the entities are selected, the processing of this flowchart ends.

[Processing based on the first standard]
The first standard is a standard using information on the usage status of the first service provided by the network NW. The first service is, for example, a service that provides an encyclopedia that can be edited by a user.

For example, the collecting unit 54 collects the content corresponding to the location information associated with the entity selected in S10 of FIG. If it has already been collected and stored in the storage unit 70, the information is referred to. For example, the location information associated with the entity of the class to be processed based on the first criterion includes, for example, the location information indicating the contents of the encyclopedia. The collection unit 54 collects data to be collected, including HTML (Hyper Text Markup Language) data of the contents of the encyclopedia and the number of accesses to the contents of the encyclopedia, from the crawl target device 30 via the network NW. ..

As the first criterion, the evaluation unit 58 links, for example, the number of accesses (or clicks) to the content of the encyclopedia and the keyword indicating the entity associated with the keyword among the keywords included in the content of the encyclopedia. The score is derived based on the probability of being included as the text (anchor text) indicating the destination. Not limited to this, the evaluation unit 58 may derive the score based on various information regarding the contents of the encyclopedia as the first criterion.

[Processing based on the second standard]
The second standard is a standard using at least information on the usage status of the second service provided by the network NW. The second service is, for example, a service different from the service that provides an encyclopedia that can be edited by a user. The second service is, for example, a service for selling a product or service (hereinafter, product, etc.) to a user, or a service for providing information on evaluation of a product or the like corresponding to an entity (for example, a service for providing information on a movie). ..

The evaluation unit 58 derives a score for each combination of the entity and the target keyword. The combination of the entity and the target keyword is the combination of the entity and the target keyword in the second score information 76. FIG. 5 is a diagram showing an example of the contents included in the second score information 76 before the score is given. The second score information 76 is information in which a keyword presumed to be related to the entity is associated with the entity. For example, the keyword associated with the entity "CDE Manga" includes a part or all of "CDE Manga" or is recognized in society as "CD-". It may be the name, abbreviation, or alias of "E-manga".

The evaluation unit 58 derives the score based on the information regarding the usage status of the second service. Therefore, the collecting unit 54 collects information on the usage status of the second service. The usage status of the second service is the number of accesses to the specific content in the second service, or the number of clicks for transitioning the screen to the specific content. The specific content is content that includes information about a product or the like corresponding to an entity included in the second service. As shown in FIG. 6, a plurality of location information may be associated with the entity to which the second criterion is applied. The location information includes, for example, location information that is different from the location information of at least the contents of the encyclopedia. That is, the location information of the specific content is included.

For example, when the entity is a "CDE cartoon", the number of accesses to a specific content (for example, a product detail page) that sells the CDE cartoon is used for deriving the score. Further, for example, when the entity is a "CDE movie", the number of accesses to the specific content evaluating the CDE movie is used for deriving the score. The evaluation unit 58 statistically processes the number of accesses or the number of clicks in a predetermined service to derive a score. The "click" is an example of selection, and may be selected by touch, tap, or other modes.

(Method of deriving the score)
The evaluation unit 58 derives the score based on at least the number of accesses to the specific content. The evaluation unit 58 may further derive the score by adding the number of accesses to the contents of the encyclopedia. In the following description, an example using the number of accesses will be described, but the number of clicks may be used instead (or in addition).

(Example 1)
For example, the evaluation unit 58 derives a score based on the number of accesses to the contents of the encyclopedia and the number of accesses to the specific contents, whichever is greater. Further, the evaluation unit 58 may derive a score based on the number of accesses obtained by adding the number of accesses to the contents of the encyclopedia and the number of accesses to the specific contents. In these cases, the higher the number of accesses, the higher the score is derived.

(Example 2)
For example, the evaluation unit 58 may derive a score based on information of a plurality of specific contents. For example, the evaluation unit 58 derives a score based on the total number of accesses to a plurality of specific contents. In this case, the higher the number of accesses, the higher the score is derived. FIG. 7 is a diagram showing an example of the content of the information on the number of accesses to a plurality of specific contents. The evaluation unit 58 may derive the score by excluding the number of accesses of the specific content having a small (or high) number of accesses from the plurality of specific contents.

(Example 3)
An index indicating accessibility may be set for each specific content, and the index may be used for deriving the score. For example, if there is a first specific content and a second specific content, and the first specific content is more accessible than the second specific content, the number of accesses to the second specific content is the second. The number of accesses may be corrected by multiplying by a coefficient that negates the inaccessibility of the specific content of. The first specific content is, for example, the content of the A shopping site, and the second specific content is, for example, the content of the B shopping site.

For example, if the accessibility of the first specific content is about half the accessibility of the second specific content, the number of accesses of the first specific content is doubled and used for deriving the score. ..

The evaluation unit 58 derives a coefficient and a score using information such as the number of accesses and the number of clicks that can be acquired in the network NW. The coefficient is the ratio of the number of accesses to the first service providing the first specific content to the number of accesses to the first specific content, and the number of accesses to the second service providing the second specific content. And may be derived based on the ratio of the number of accesses to the second specific content.

By performing the above-mentioned processing for each entity, a score is given to the combination of the entity and the keyword in the second score information 76. Then, the second score information 76 to which the score is given is stored in the storage unit 70.

FIG. 8 is a diagram showing an example of the contents of the second score information 76 to which a score is given. The score is given for each combination of the entity and the keyword. For example, when the same keyword is associated with a plurality of entities, a score is given for each combination. By giving a score in this way, for example, when a query associated with a plurality of entities is input and an entity for the query is selected, the query (keyword) is as described with reference to FIG. 9 described later. The appropriate entity is selected by referring to the score given to the combination of and the entity.

[Providing a knowledge panel]
The response unit 60 acquires a query from the search server 20 and provides the search server 20 with information corresponding to the acquired query. FIG. 9 is a flowchart showing an example of the flow of processing for providing the knowledge panel to the user. First, the response unit 60 determines whether or not a query has been received from the search server 20 (S100). Upon receiving the query, the response unit 60 refers to the first score information 74 and the second score information 76 to which the score is given, extracts the entity associated with the keyword matching the query as an entity candidate, and extracts the entity. Further, the score of the entity candidate is acquired (S102).

Next, the response unit 60 determines the entity candidate to which the highest score is given among the entity candidates as an entity (S104). Next, the response unit 60 transmits the location information of the entity associated with the keyword in the knowledge graph 72 to the search server 20 (S106). In S104, when the entity candidate with the score equal to or higher than the threshold value does not exist, the response unit 60 does not transmit the location information or transmits information indicating that there is no entity satisfying the criteria to the search server 20. You may do it. This ends the processing of this flowchart.

FIG. 10 is a conceptual diagram of a process flow for providing a knowledge panel to a user. For example, when the query "CDE" is entered, each of the entities "CDE cartoon", "CDE animation", "CDE movie" and the keyword "CDE movie" are entered. Among the scores given to the combination with "CDE", the entity "CDE cartoon" having the highest score is determined as the entity for the input query. Then, the content as shown in FIG. 11 is provided to the user who has input the query.

[Example of knowledge panel]
FIG. 11 is a diagram showing an example of a knowledge panel provided to a user. The display unit 12 of the terminal device 10 shown in FIG. 11 displays the content including the query input area 13, the knowledge panel 14, and the search result 15. The knowledge panel 14 includes the entity 16 corresponding to the query, the image 17 related to the entity 16, and the related information 18 related to the entity 16. The search server 20 extracts information related to the entity from the contents indicated by the location information transmitted by the knowledge data server 50. The search server 20 generates content including information related to the extracted entity (for example, information included in the knowledge panel 14), and provides the generated content to the user's terminal device 10. As a result, the information shown in FIG. 11 is displayed on the display unit 12 of the terminal device 10.

[Comparison]
For example, the knowledge data server of the comparative example may perform processing by uniformly applying the "first standard" without applying the "second standard". For example, there are manga, anime, movies, novels, etc. based on the character "CDE", and the entities are "CDE manga", "CDE animation", and "C-". It is assumed that "DE movie" and "CDE novel" exist. In this case, if the query "CDE" is entered, the entity is selected based on the information in the encyclopedia content, so it is not possible to provide the information that many users think they want. There was a weak connection between the information in the encyclopedia content and the entity that many users would want, and conversely the strong connection with the entity that a few users would want. Also, it does not take into account how the concepts of each entity are used in society, whether they are pervasive, or whether they are popular.

On the other hand, the knowledge data server 50 of the present embodiment applies the "second criterion" to a predetermined entity, gives a score to the combination of the entity and the keyword, and based on this score. By selecting an entity, it is possible to provide the information that many users think they want. For example, in the second criterion, when a score is given, how the concept of each entity is used, permeated, or popular in society is added more positively. Therefore, it is possible to provide the user with the information that the user wants more.

According to the first embodiment described above, when the knowledge data server 50 refers to the attribute information associated with the entity and determines that the entity is the target of application of the second criterion, it is based on the second criterion. , By evaluating the combination of the entity and the keyword, the relationship between the entity and the keyword can be appropriately evaluated.

<Second Embodiment>
Hereinafter, the second embodiment will be described. In the second embodiment, in the process of applying the second criterion, auxiliary information is used for deriving the score in addition to the information about the specific content.
Hereinafter, the differences from the first embodiment will be described.

The auxiliary information is, for example, information regarding the degree of interest of an entity in society (for example, information indicating a variable factor of the degree of interest). Fluctuation factors include, for example, an index showing popularity, the number of copies of a book, the number of spectators in a movie, the date of publication of a book, the date of broadcasting an animation (or TV drama), the date of showing a movie, and the like. The index indicating the popularity is, for example, an index (correction value) derived based on the number of clicks on the information contained in the knowledge panel 14. For example, the knowledge panel 14 contains information indicating "CDE cartoon", information indicating "CDE animation", information indicating "CDE movie", and "C". When the number of clicks on the information indicating "DE movie" is higher than the number of clicks on other information, the popularity of "CDE movie" is "CDE cartoon" and "CD". -It is considered to be higher than "E Anime". in this case. A large correction value is set for the entity corresponding to the information with a high number of clicks.

In addition, if the circulation of manga is more than the specified number, or if the manga is issued within the specified period, the popularity of "CDE Manga" is set to a higher correction value than other entities. May be done. Further, when the animation is broadcast within a predetermined period, the popularity of the "CDE animation" may be set to a higher correction value than that of other entities.

FIG. 12 is a diagram showing an example of the contents of the second score information 76A to which the auxiliary information is added.
The second score information 76A is associated with, for example, a correction value in addition to the content of the second score information 76. This correction value is associated with each entity, for example. For example, if the "CDE movie" is presumed to be more interested than the other entities, the "CDE movie" is given a higher correction value than the other entities. There is. Then, the derived score is corrected by the correction value, and the corrected score becomes a comparison target when the entity is selected.

According to the second embodiment described above, since the score is adjusted based on the variable factor of the degree of interest of the entity, the relationship between the entity and the keyword can be evaluated more appropriately. As a result, a more appropriate entity is selected and more appropriate information is provided to the user.

<Third Embodiment>
Hereinafter, the third embodiment will be described. In the first embodiment, the evaluation unit 58 applies the first criterion or the second criterion based on the class of the entity to derive the score. On the other hand, in the second embodiment, the evaluation unit 58 derives the score based on the criteria associated with the class of the entity. Hereinafter, the differences from the first embodiment will be described.

The evaluation unit 58 derives a score based on, for example, the reference information 78 stored in the storage unit 70. FIG. 13 is a diagram showing an example of the contents of the reference information 78. The reference information 78 is information in which the class and the criteria for deriving the score are associated with each other. The criteria for deriving the score are, for example, the criteria for using the number of accesses to the contents of the encyclopedia, the criteria for using the number of accesses to the contents of the encyclopedia and the contents of a predetermined shopping site, and the criteria for the number of accesses to the contents of multiple shopping sites. It is a standard etc. to use. For example, the evaluation unit 58 refers to the reference information 78, extracts the criteria corresponding to the class associated with the entity, and derives the score based on the information according to the extracted criteria.

According to the third embodiment described above, since the score is derived based on the criteria associated with each class, the relationship between the entity and the keyword can be evaluated more appropriately. As a result, a more appropriate entity is selected and more appropriate information is provided to the user.

In each of the embodiments described above, the determination unit 56 that refers to the attribute information associated with the entity and determines whether the entity is the application target of the first criterion or the entity is the application target of the second criterion. , If it is determined that the first criterion is applied, the combination of the entity and the keyword is evaluated based on the first criterion, and if it is determined that the second criterion is applied, the second criterion is applied. By providing the evaluation unit 58 that evaluates the combination based on the above, the relationship between the entity and the keyword can be appropriately evaluated. As a result, the appropriate entity can be selected and provide the user with more appropriate information.

Although the embodiments for carrying out the present invention have been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. Can be added.

10 ... Terminal device, 20 ... Search server, 30 ... Crawling target device, 50 ... Knowledge data server, 52 ... Communication unit, 54 ... Collection unit, 56 ... Judgment unit, 58 ... Evaluation unit, 60 ... Response unit, 70 ... Memory Department, 72 ... Knowledge Graph, 74 ... 1st Score Information, 76 ... 2nd Score Information

Claims (10)

A determination unit that refers to the attribute information associated with the entity and determines whether the entity is the first type or the entity is the second type .
If it is determined to be the first type, the combination of the entity and the query is evaluated based on the information about the first content of the encyclopedia corresponding to the entity provided by the network and editable by the user. ,
When it is determined to be the second type, the usage status of the second content, which is the content for selling the product or service (hereinafter, product, etc.) related to the entity, or the content for providing the user with information regarding the evaluation of the product, etc. Based on, the above combination is evaluated and
An evaluation unit that generates information for determining an entity according to an input query using the evaluation result, and an evaluation unit.
Information processing device equipped with. When the evaluation unit determines that the second type is used, the number of accesses to the second content, the number of clicks for transitioning the screen to the second content, or the number of clicks is statistically processed. Based on the result, the above combination is evaluated.
The information processing apparatus according to claim 1. When the evaluation unit is determined to be the second type, the evaluation unit further evaluates the combination in consideration of the date when the medium corresponding to the entity is released, aired, or screened.
The information processing apparatus according to claim 2. The entity to which the attribute information determined to be the second type is associated includes the concept commonly used by the same expression in a plurality of media.
The information processing apparatus according to any one of claims 1 to 3. The media may be cartoons, anime, movies, television, novels, or music.
The information processing apparatus according to claim 4. It further includes a response unit that selects an entity based on the input query and the evaluation result of the evaluation unit and provides the user with information about the selected entity.
The information processing apparatus according to any one of claims 1 to 5. The evaluation unit
When it is determined to be the first type, in the first content of the encyclopedia corresponding to the entity provided by the network and editable by the user, the target query associated with the entity and indicating the entity links to the link destination. A first index for the combination is derived based on the degree to which it is included as the indicated anchor text.
When it is determined that the second type is used, a second index for the combination is derived based on at least the usage status of the second content.
Generates information to determine the entity according to the entered query based on the first metric or the second metric.
The information processing apparatus according to any one of claims 1 to 6. A provision unit that provides the user with a third content associated with the entity determined based on the information for determining the entity.
The information processing apparatus according to any one of claims 1 to 7. The computer
With reference to the attribute information associated with the entity, it is determined whether the entity is the first type or the entity is the second type .
If it is determined to be the first type, the combination of the entity and the query is evaluated based on the information about the first content of the encyclopedia corresponding to the entity provided by the network and editable by the user. ,
When it is determined to be the second type, the usage status of the second content, which is the content for selling the product or service (hereinafter, product, etc.) related to the entity, or the content for providing the user with information regarding the evaluation of the product, etc. Based on, the above combination is evaluated and
Using the result of the evaluation, information for determining the entity according to the input query is generated.
Information processing method. On the computer
With reference to the attribute information associated with the entity, it is determined whether the entity is the first type or the entity is the second type .
If it is determined to be the first type, the combination of the entity and the query is evaluated based on the information about the first content of the encyclopedia corresponding to the entity provided by the network and editable by the user. Let me,
When it is determined to be the second type, the usage status of the second content, which is the content for selling the product or service (hereinafter, product, etc.) related to the entity, or the content for providing the user with information regarding the evaluation of the product, etc. To evaluate the combination based on
Using the result of the evaluation, the information for determining the entity according to the input query is generated.
program.

Images