JP6971104B2 - 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.

Conventionally, a crawler that collects data (documents, images, etc.) from the Web and automatically creates a database of the collected data is known (see Patent Document 1). Crawlers collect data from web pages with various IP addresses by following links in web pages. The data collected by the crawler is stored in the web information database.

On the other hand, when a search engine receives a search word input by a user, it acquires information on a web page or application page related to the received search word (for example, URL: Uniform Resource Locator) from a database, and a list of the acquired information. Is output as a search result.

Japanese Unexamined Patent Publication No. 2012-69171

A search engine may be operated to output data related to a search word (query) input by a user together with a search result in order to improve user satisfaction. In order to achieve this, a database in which text data and entities are associated with each other and stored may be used. Here, when text data associated only with a specific entity is input as a query in the database, that specific entity is naturally output together with the search result. However, if this text data is incorrect or inappropriate as information about the entity, inappropriate information may be output together with the search results.

The present invention has been made in consideration of such circumstances, and one of the objects of the present invention is to provide an information processing apparatus capable of suppressing output of inappropriate information together with a search result.

One aspect of the present invention is a selection unit that selects an arbitrary target text data from the content information associated with the entity and the text data, and one or more related to the target text data selected by the selection unit. In the collection unit that collects related web pages, and in the first type of related web page among one or more related web pages collected by the collection unit for each entity associated with the target text data in the content information. The first probability that the target text data is included as the text indicating the link destination, and the entity associated with the target text data in the content information in the second type of related web page among the one or more related web pages. Of the second probability that the link destination to be included as text is selected, at least one and the web page at the link destination indicated by the text in the first type related web page or the second type related web page. An information processing apparatus including an evaluation unit that evaluates the content information as an entity corresponding to the target text data based on a third probability that is a web page of the entity associated with the target text data. be.

According to one aspect of the present invention, it is possible to suppress the output of inappropriate information together with the search result.

It is a figure which shows the use environment and the configuration of the knowledge data server which concerns on this embodiment. It is a figure which shows an example of the knowledge graph which concerns on this embodiment. It is a figure which shows an example of the web page of "baseball player A" which concerns on this embodiment. It is a figure which shows an example of the web page of "baseball team B" which concerns on this embodiment. It is a figure which shows an example of the content information which concerns on this embodiment. It is a figure which shows an example of the search result web page which concerns on this embodiment. It is a figure which shows another example of the content information which concerns on this embodiment. It is a figure which shows another example of the search result web page which concerns on this embodiment. It is a figure which shows an example of the related web page collected by the collection part which concerns on this embodiment. It is a figure which shows another example of the related web page collected by the collection part which concerns on this embodiment. It is a figure which shows an example of the same entity which concerns on this embodiment. It is a figure which shows another example of the web page of "baseball player A" which concerns on this embodiment. It is a figure which shows an example of the content information and the score which concerns on this embodiment. It is a flow chart which shows an example of the operation which calculates the score of the knowledge data server of this embodiment. It is a flow chart which shows an example of the operation based on the knowledge data server score of this embodiment.

Hereinafter, embodiments of the information processing apparatus, information processing method, and program of the present invention will be described with reference to the drawings. In the present embodiment, the information processing apparatus will be described as forming a part of the knowledge data server. The knowledge data server is, for example, a server that collects data to be collected (for example, image or text data) and generates a knowledge graph based on the collected data. In the present embodiment, the knowledge graph is data in which an entity, a class, and a property are associated with the location information of a web page in which related data related to the entity is described. 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, defined in the real world or virtual world). It may represent a concept). 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". Hereinafter, embodiments will be described in detail.

<Embodiment>
<1-1. Knowledge data server usage environment>
FIG. 1 is a diagram showing a usage environment and configuration of the knowledge data server 100 according to the present embodiment. The knowledge data server 100 is communicably connected to the terminal device 200, the web server 300, the crawl target device DV, and the network NW. The network NW means the World Wide Web, and is a system using HTML documents and the like that are standardly used on the Internet and intranets. The network NW may further include a radio base station, a provider device, a dedicated line, and the like.

The terminal device 200 is a device used by a user, and is, for example, a mobile phone such as a smartphone, a tablet computer, a notebook computer, a desktop computer, or the like. The web server 300 is a server that uses a search engine to generate a web page for displaying search results and provides it to the terminal device 200.

The knowledge data server 100 includes, for example, a control unit 110 and a storage unit 120. The storage unit 120 is realized by, for example, a RAM (Random Access Memory), an HDD (Hard Disk Drive), a flash memory, or a hybrid storage device in which a plurality of these are combined. Further, a part or all of the storage unit 120 may be an external device such as NAS (Network Attached Storage) or an external storage server that can be accessed by the knowledge data server 100. The knowledge graph (hereinafter referred to as knowledge graph D1) and the content information D2 are stored in the storage unit 120.

FIG. 2 is a diagram showing an example of Knowledge Graph D1 according to the present embodiment. The things described in Knowledge Graph D1 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.

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 entities, or things, shall belong to any of the classes contained within 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 D1 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.

In the knowledge graph D1 of the example shown in FIG. 2, the entity E1 and the entity E2 belong to a class having a property name of "belonging team". In the present embodiment, each entity is associated with information that can identify each entity (hereinafter, entity identification information EID), an entity name, and location information of a web page of the entity. The location information is information for specifying a position on the Web, and is, for example, a URL (Uniform Resource Locator). The web page indicated by the location information contains the relevant data of the entity. In the knowledge graph D1 shown in FIG. 2, the entity E1 is the location of a web page in which the entity identification information EID "0001", the entity name "baseball player A", and related data related to the entity are described. It is associated with the information "http: // encyclopedia web page / baseball player A" URL. Further, in the entity E2, the entity identification information EID "0002", the entity name "baseball team B", and the location information of the web page in which the related data related to the entity are described are described as "http: //". It is associated with the URL "Encyclopedia Web Page / Baseball Team B". In the following description, the related data related to the entity will be described as the related data of the entity. In addition, the web page in which the data related to the entity is described is also described as the web page of the entity.

FIG. 3 is a diagram showing an example of a web page of “baseball player A” according to the present embodiment. The web page of "baseball player A" contains data related to "baseball player A". Further, FIG. 4 is a diagram showing an example of a web page of “baseball team B” according to the present embodiment. The web page of "Baseball Team B" contains data related to "Baseball Team B".

FIG. 5 is a diagram showing an example of the content information D2 according to the present embodiment. The content information D2 is information in which an entity and text data are associated with each other. The content information D2 is generated based on the collection result of collecting the related data contained in the web page or the search log of the search engine. The content information D2 is, for example, information in which text data included in a web page of an entity is collected and the collected text data is associated with the entity. Further, the content information D2 is, for example, information in which, when an entity is input as a query, the text data input together with the entity is associated with the entity. In the example content information D2 shown in FIG. 5, the entity "baseball player A" is associated with the text data "3000 hits", and the entity "baseball team B" is referred to as "2017 championship". Text data is associated.

<1-2. Search results web page ＞
FIG. 6 is a diagram showing an example of a search result web page according to the present embodiment. As shown in FIG. 6, the display unit 210 displays a search result web page including a query input area 211, a knowledge panel 220, and a search result 230. The knowledge panel 220 includes the entity 221 corresponding to the query, the image 222 associated with the entity 221 and the relevant information 223 associated with the entity 221. The user uses the terminal device 200 to input a query on the search page displayed on the display unit 210 of the terminal device 200. A query is a search word or a combination of multiple search words. The terminal device 200 sends the query entered by the user to the web server 300.

The web server 300 transmits the query received from the terminal device 200 to the knowledge data server 100. The knowledge data server 100 sends information related to the entity corresponding to the received query to the web server 300. The knowledge data server 100 determines an entity associated with a received query (text data), for example, based on the content information D2. Further, the knowledge data server 100 transmits the location information associated with the determined entity to the web server 300 based on the knowledge graph D1. Based on the received location information, the web server 300 extracts the relevant data of the entity from the web pages indicated by the location information. The web server 300 generates a search result web page (eg, FIG. 6) that includes relevant data of the extracted entity (in this example, the information shown in the knowledge panel 220).

In the example shown in FIG. 6, "3000 hits" is input as a query, and an image and various information related to "3000 hits" are displayed on the knowledge panel 220. On the knowledge panel 220, for example, an image 222 of "baseball player A" who achieved "3000 hits" and related information 223 such as the date of birth and birthplace of "baseball player A" are displayed. In this way, by displaying not only the search result 230 but also the knowledge panel 220 on the display unit 210, the user satisfaction with the search result can be improved.

<1-3. Content information D2 associated with minority text data>
FIG. 7 is a diagram showing content information D2a, which is another example of content information D2 according to the present embodiment. Here, in the content information D2, the same text data may be associated with a plurality of different entities. For example, an entity indicating a person is associated with "male" text data or "female" text data indicating the gender of the person. Therefore, even if the entities indicate different persons, the same text data (for example, "male" or "female") is associated with each other. On the other hand, an entity may be associated with text data that is not often associated with other entities. FIG. 7 is a diagram showing content information D2 (hereinafter, content information D2a) in which a minority text data is associated with an entity.

In the example shown in FIG. 7, in the content information D2a, the text data of the minority "violent man" is associated with the entity "baseball player A". Here, at the time of generating the content information D2, it is not verified whether the text data associated with the entity is a general opinion (majority) or a minority opinion. For example, if there is a web page in the network NW that describes the content of a minority (for example, "violent man") for "baseball player A", the content as shown in FIG. 7 based on the web page. Information D2a is generated.

<1-4. Knowledge Panel 220 Based on Minority Opinions>
FIG. 8 is a diagram showing another example of the search result web page according to the present embodiment. As text data, which is a minority opinion, is input to the query input area 211, the display unit 210 displays information on the entity associated with the text data, for example, on the knowledge panel 220. The result web page is displayed. In the example shown in FIG. 8, when the text data "violent man" is input to the query input area 211, in the knowledge data server 100, the entity associated with the "violent man" is "baseball player A". Is determined to be. Then, the knowledge data server 100 supplies the location information of the web page of the "baseball player A" to the web server 300 as the information of the entity corresponding to the query. The web server 300 displays a knowledge panel 220 including an image 222 of "baseball player A" and related information 223 such as the date of birth and birthplace of "baseball player A" as related data related to "violent man". Generate a search result web page. In this case, the user refers to the search result web page and gets the impression that "baseball player A" is a "violent man". In addition, if a malicious user wants to give a bad impression that "baseball player A" is a "violent man", he / she can describe the content that "baseball player A" is a "violent man" on the web page. It is possible to induce the knowledge graph D1 to register this relationship and display the search result web page as shown in FIG.

In order to suppress such processing, it is preferable that the text data associated with the entity obtains an index indicating whether the text data is a majority or a minority. The knowledge data server 100 of the present embodiment derives an index (hereinafter referred to as a score) indicating whether the text data associated with the entity is appropriate. Hereinafter, the detailed configuration of the knowledge data server 100 will be described.

<1-5. Configuration of Knowledge Data Server 100>
Returning to FIG. 1, the control unit 110 is realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program (software). The control unit 110 includes, for example, a selection unit 111, a collection unit 112, a score calculation unit 113, a location information acquisition unit 114, and a communication I / F (Interface) 115. In addition, some or all of these components (excluding the contained storage unit) are LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing). It may be realized by hardware (circuit unit; including circuitry) such as Unit), or it may be realized by the cooperation of software and hardware. The communication I / F 115 mediates communication between each functional unit via the network NW and other devices (for example, a crawl target device (hereinafter, crawl target device DV), a terminal device 200, and a web server 300). ..

The selection unit 111 selects arbitrary text data (hereinafter referred to as target text data) to be scored from the content information D2. The collection unit 112 collects a web page (hereinafter referred to as a related web page) related to the target text data selected by the selection unit 111 via the network NW. The web page collected by the collection unit 112 is, for example, an encyclopedia web page that can be edited by a user, or a search result web page when the target text data is used as a query. Hereinafter, the encyclopedia web page that can be edited by the user is referred to as an encyclopedia web page. Here, the encyclopedia web page is an example of the first type of related web page. The search result web page is an example of a second type of related web page.

FIG. 9 is a diagram showing an example of an encyclopedia web page collected by the collection unit 112. In the example shown in FIG. 9, the collecting unit 112 describes the target text data of "3000 hits" associated with the entity of "baseball player A" indicated by the content information D2 in the encyclopedia web page. Collect web pages. The collection unit 112 collects data to be collected including HTML (HyperText Markup Language) data of an encyclopedia web page from a crawl target device DV via a network NW, for example. The collecting unit 112 is realized by a so-called crawler program. Here, the data to be collected is data that is on the network NW (in the storage area of the device DV to be crawled) and can be viewed by a browser. The data to be collected is not limited to the browser, and may be data reproduced by the application program. The network NW means the World Wide Web, and is a system using HTML documents and the like that are standardly used on the Internet and intranets. The data to be collected is, for example, a related web page containing text data indicating the source of HTML. The collection unit 112 collects all the related web pages in which the target text data is described among the encyclopedia web pages.

Further, FIG. 10 is a diagram showing an example of a search result web page collected by the collection unit 112. In the example shown in FIG. 10, in the search result web page, a query (illustrated text TX21) in which the entity "baseball player A" indicated by the content information D2 is displayed as a search result is input. The search results are shown. The query is an example of target text data.

Returning to FIG. 1, the score calculation unit 113 calculates the score of the set of the target text data and the entity based on the related web page collected by the collection unit 112. The collecting unit 112 calculates, for example, the value of P (e | s) represented by the equation (1) as a score. Various elements in the formula (1) will be described for each of the formulas (2) and (3) described later and the types of related web pages.

Hereinafter, the details of the score calculation process when the related web page is an encyclopedia web page will be described, and then the details of the score calculation process when the related web page is a search result web page will be described.

<1-6. If the related web page is an encyclopedia web page>
The score calculation unit 113 calculates the probability that the text data indicating the entity associated with the target text data is included as the text (anchor text) indicating the link destination among the text data included in the encyclopedia web page. The related web page contains text data (illustrated text TX12) indicating the target text data (in this example, "3000 hits" (illustrated text TX11) and the entity associated with it (in this example, "baseball player A")). ~ 14) is included. Further, among the text TX12-14, the text TX12 is an anchor text. Further, the link of another web page given to the text TX12 is included in the target text data in the knowledge graph D1. The location information associated with the associated entity (in this example, "baseball player A").

Here, the score calculation unit 113 is a probability obtained by the number of target text data which is an anchor text with respect to the number of target texts in all the encyclopedia web pages including the target text data by the following formula (2). (Hereinafter, anchor text probability) is calculated. _{A s} in equation (2) shows the anchor text probability. Here, the anchor text probability is an example of the first probability.

Further, the score calculation unit 113 uses the following formula (3) to indicate that the link destination of the anchor text is the web page of the entity corresponding to the location information associated with the target text data in the knowledge graph D1 (hereinafter, specific web). Page) is calculated (hereinafter referred to as the first entity probability). The e in the equation (3) indicates the first entity probability. Here, the first entity probability is an example of the third probability.

When the related web page collected by the collecting unit 112 is only the web page shown in FIG. 9, the score calculation unit 113 uses the target text data as the anchor text among the target text data included in all the related web pages. The probability (that is, the anchor text probability) is calculated as "1/3". This is because, in the target text data shown in the figure, only one link is given (even if two or more links are given, it may be counted as one). Further, the score calculation unit 113 calculates the probability (first entity probability) that the link destination of the anchor text is a specific web page of the entity associated with the target text data as "1".

<1-7. If the related web page is a search result web page>
The score calculation unit 113 calculates the probability that the link destination including the entity associated with the target text data included in the search result web page is selected as the text. The target text data is "3000 hits" (text TX21 in the figure), and the entity associated with the target text data is "baseball player A". In the following description, the text related to the entity associated with the target text data will be referred to as the target entity text data. In the example of FIG. 10, the related web page includes text data (texts TX22 to 24 shown) to which a link of the related web page of the target text data is added. Of these, the text TX22 contains the target entity text data (in this example, "baseball player A"). The link destination given to the text TX 22 is, for example, the location information associated with the entity (“baseball player A” in this example) associated with the target text data in the Knowledge Graph D1. That is, the link of another web page of the text TX22 is a specific web page, for example, an encyclopedia web page. It should be noted that the search result web page may include text TX25, which is text without a link and indicates an entity.

The score calculation unit 113 determines the ratio of the number of times the link destination including the target entity text data is clicked to the number of times each link destination included in all the search result web pages is clicked by the formula (2). Calculate the required probability (hereinafter, entity text click probability). The "click" is an example of selection, and may be selected by touch, tap, or other modes. In the following description, it is collectively referred to as click. A _s in equation (2) shows the entity text click probability. Here, the entity text click probability is an example of the second probability.

Further, the score calculation unit 113 calculates the probability that the link destination including the target entity text data is a specific web page related to the entity (hereinafter referred to as the second entity probability) by the equation (3). E in the equation (3) indicates the second entity probability. Here, the second entity probability is another example of the third probability.

<1-8. When the same entity pointing to different things is included>
Here, an entity may refer to different things even if they have the same wording. FIG. 11 is a diagram showing an example of linking to a web page of a different entity by the same wording. In the example shown in FIG. 11, the text TX22 indicates a link destination of a web page (for example, FIG. 3) of "baseball player A". On the other hand, the text TX26 indicates a link destination of a web page of a player who is a baseball player with the same surname and the same name as "baseball player A" and belongs to a baseball team C different from "baseball player A". The score calculation unit 113 also includes the click of the text TX26 in the entity click probability in the equation (2).

FIG. 12 is a diagram showing an example of a web page relating to another entity linked from the wording indicating the entity. The web page of "baseball player A" shown in FIG. 12 is a baseball player different from "baseball player A" ("baseball player A" shown in FIG. 3) associated with the text data "3000 hits". Related data related to a certain "baseball player A" is described.

When the score calculation unit 113 calculates the probability of the first entity, if the link destination is the web page shown in FIG. 12, the score calculation unit 113 does not count the web page as the web page of the entity.
Further, even when the score calculation unit 113 calculates the second entity probability, if the link destination is the web page shown in FIG. 12, the score calculation unit 113 does not count the web page as the web page of the entity.

<1-9. Weighting for each related web page>
Further, the score calculation unit 113 further multiplies the value obtained by multiplying the anchor text probability calculated when the related web page is the encyclopedia web page by the first entity probability by the following equation (4). You may. Further, the score calculation unit 113 further applies the following equation (4) to the value obtained by multiplying the entity text click probability calculated when the related web page is the search result web page by the second entity probability. You may multiply. Then, the score calculation unit 113 calculates the score by, for example, adding both to obtain a logarithm.

Equation (4) is the number of samples for which equations (2) and (3) are calculated using the encyclopedia web page, and the number of samples for which equations (2) and (3) are calculated using the search result web page. The ratio with and is shown.

The score calculation unit 113 further multiplies the ratio of the number of samples for which equations (2) and (3) are calculated using the encyclopedia web page by the value obtained by multiplying the anchor text probability and the first entity probability, and searches. By further multiplying the ratio of the number of samples calculated in equations (2) and (3) using the result web page by the value obtained by multiplying the entity text click probability and the second entity probability, and adding both. Weight the score.

The score calculation unit 113 stores the calculated score in association with the content information D2. FIG. 13 is a diagram showing an example of the content information D2 and the score according to the present embodiment.

Returning to FIG. 1, the location information acquisition unit 114 refers to the content information D2 stored in the storage unit 120, and acquires an entity corresponding to the query received from the web server 300. When the text data matching the received query is associated with a score equal to or higher than a predetermined threshold value, the location information acquisition unit 114 determines the entity associated with the text data as the entity corresponding to the query. do. The location information acquisition unit 114 acquires the location information associated with the determined entity based on the knowledge graph D1. After that, the location information acquisition unit 114 transmits the acquired location information to the web server 300. Here, the location information acquisition unit 114 is an example of a determination unit.

In this way, when the location information acquisition unit 114 receives the query from the web server 300, the location information acquisition unit 114 identifies the text data corresponding to the query. At this time, not only when the query and the text data are exactly the same, but also when there is a partial match or a difference to the extent that they are considered to be semantically the same, it may be regarded as the corresponding text data.

<1-10. Operation of Knowledge Data Server 100>
Hereinafter, the operation of the knowledge data server 100 will be described. FIG. 14 is a flow chart showing an example of an operation of calculating the score of the knowledge data server 100 of the present embodiment. The selection unit 111 selects the target text data to be scored from the content information D2 (S110). Next, the collection unit 112 collects related web pages (encyclopedia web page and search result web page) based on the target text data (S120). Next, the score calculation unit 113 calculates the anchor text probability and the entity text click probability (S130).

Next, the score calculation unit 113 calculates the first entity probability and the second entity probability (S140). Then, the score calculation unit 113 multiplies the anchor text probability based on the encyclopedia web page, the first entity probability, and the first content ratio, the entity text click probability based on the search result web page, the second entity probability, and the like. And, the multiplication with the value obtained by multiplying the second content ratio is calculated as a score (S150), and the calculated score is associated with the content information D2 and stored in the storage unit 120 (S160).

FIG. 15 is a flow chart showing an example of an operation based on the score of the knowledge data server 100 of the present embodiment. The location information acquisition unit 114 receives a query from the web server 300 (S210). Next, the location information acquisition unit 114 reads out the score associated with the text data matching the query in the content information D2 (S220). Next, the location information acquisition unit 114 determines whether the score is equal to or higher than a predetermined threshold value (S230). When the score is equal to or higher than a predetermined threshold value, the location information acquisition unit 114 transmits the location information of the entity associated with the text data in the knowledge graph D1 to the web server 300 (S240). Further, the location information acquisition unit 114 does not transmit the location information to the web server 300 when the specification is less than a predetermined threshold value (S250).

<1-11. Variations in score calculation>
In the above description, the case where the score calculation unit 113 calculates the score based on the anchor text probability, the entity text click probability, the first entity probability, the second entity probability, the first content ratio, and the second content ratio will be described. However, it is not limited to this. For example, if the number of encyclopedia web pages collected is small among the related web pages, various values related to the encyclopedia web pages (for example, anchor text probability, first entity probability, and first content ratio) are used to calculate the score. The impact on may be small. Therefore, the score calculation unit 113 does not have to use the anchor text probability, the first entity probability, and the first content ratio to calculate the score. In addition, when the number of collected search result web pages is small among the related web pages, various values related to the search result web pages (for example, entity text click probability, second entity probability, and second content ratio) are used to calculate the score. The impact on may be small. Therefore, the score calculation unit 113 does not have to use the entity text click probability, the second entity probability, and the second content ratio to calculate the score.

<1-12. Evaluation method other than score>
Further, in the above description, the case where the text data of the content information D2 is evaluated by the score calculated by the score calculation unit 113 has been described, but the present invention is not limited to this. The score calculation unit 113 may be configured to evaluate the target text data in two stages of “appropriate” and “inappropriate” based on the calculated score, for example. In this case, the score calculation unit 113 associates the target text data with information indicating "appropriate" when the calculated score is equal to or higher than a predetermined threshold value. Further, when the calculated score is less than a predetermined threshold value, the score calculation unit 113 associates the target text data with information indicating “inappropriate”. When the acquired query matches the text data to which "inappropriate" is associated, the location information acquisition unit 114 is "inappropriate" to supply the location information of the entity associated with the text data. It is determined that the location information is not supplied to the web server 300. The score calculation unit 113 may evaluate the target text data in three or more stages instead of the configuration in which the target text data is evaluated in two stages.

As described above, in the knowledge data server 100 of the present embodiment, the selection unit 111 and the selection unit 111 for selecting arbitrary target text data in the content information D2 in which the entity and the text data are associated with each other are used. A collection unit 112 that collects one or more related web pages related to the selected target text data, and one or more related webs collected by the collection unit 112 for each entity associated with the target text data in the content information D2. In the first type of related web page of the page, the first probability that the target text data is included as the text indicating the link destination, and in the second type of related web page of one or more related web pages, in the content information D2. A link indicated by at least one of the second probability that a link destination containing an entity associated with the target text data as text is selected and the text in the first type related web page or the second type related web page. An evaluation unit (score calculation unit 113) that evaluates as an entity corresponding to the target text data based on a third probability that the preceding web page is the web page of the entity associated with the target text data in the content information D2. ) And. As a result, the knowledge data server 100 of the present embodiment can suppress the output of inappropriate information together with the search result.

Here, in the web page existing in the network NW, there may be a web page with an explanation that is not appropriate as a word related to the entity. However, the number of web pages with inappropriate explanations may be small compared to the number of web pages with proper explanations. When the knowledge data server 100 of the present embodiment has a low appearance probability of the text data of the content information D2 on the web page (encyclopedia web page or search result web page) existing in the network NW, the text data is low. Give a score. Further, the knowledge data server 100 of the present embodiment gives a high score to the text data when the appearance probability of the text data of the content information D2 is high on the web page existing in the network NW. As a result, the knowledge data server 100 of the present embodiment can calculate the score as an index of whether or not the word is appropriate as a word related to the entity.

Further, in the knowledge data server 100 of the present embodiment, the score calculation unit 113 transitions from the search result web page to the web page of the entity associated with the target text data in the content information D2, and the content from the encyclopedia web page. The first probability and the second probability are weighted according to the ratio of the occurrence probability to the event that has transitioned to the web page of the entity associated with the target text data in the information. According to the knowledge data server 100 of the present embodiment, the score can be calculated with higher accuracy as an index of whether or not the word is appropriate as a word related to an entity.

Further, in the knowledge data server 100 of the present embodiment, the collection unit 112 collects a related web page (in this example, an encyclopedia web page) from at least an encyclopedia that can be edited by a user. Further, in the knowledge data server 100 of the present embodiment, the collecting unit 112 relates at least a web page for displaying a search result when the target text data is used as a query (in this example, a search result web page). Collect as a web page. Here, when searching for a search target existing in the network NW, if the search target range is not defined, the processing load related to the search may become enormous. According to the knowledge data server 100 of the present embodiment, an encyclopedia web page and a search result web page are collected as related web pages. Therefore, the knowledge data server 100 of the present embodiment can collect related web pages by a simple process.

Further, in the knowledge data server 100 of the present embodiment, the location information acquisition unit 114 refers to the evaluation result by the score calculation unit 113 (in this example, the score or the evaluation in two stages) with respect to the input query. Judge whether to add to the search result and output. Specifically, in the knowledge data server 100 of the present embodiment, the evaluation unit (in this example, the score calculation unit 113) calculates the score of the entity corresponding to the target text data, and the determination unit (in this example, the location). The information acquisition unit 114) outputs the information of the entity to which the score is associated when the score is equal to or higher than the predetermined threshold, and outputs the information of the entity to which the score is associated when the score is less than the predetermined threshold. Do not output. As a result, the knowledge data server 100 of the present embodiment outputs the entity-related data to the web server 300 when the query is appropriate as text data associated with the entity. Therefore, the knowledge data server 100 of the present embodiment can provide an appropriate knowledge panel 220 while improving the user satisfaction with the search result by displaying the knowledge panel 220 on the display unit 210.

<1-13. About blacklist of text data ＞
In the above description, the knowledge data server 100 has described the case where the location information of the entity is transmitted to the web server 300 when the score is equal to or higher than a predetermined threshold value, but the present invention is not limited to this. Here, in the query input area 211, a query that is not suitable for displaying the knowledge panel 220 may be input. Queries that are not suitable for displaying the knowledge panel 220 are, for example, words that defamate a person, words that disturb public order and morals, and words that detract from the reputation of a product. In this case, the score calculation unit 113 may be configured to generate information (hereinafter, text data blacklist) indicating the content information D2 that does not provide the location information of the entity to the web server 300 among the content information D2. The score calculation unit 113 refers to a dictionary such as a word that defames a person, a word that disturbs public order and morals, and a word that detracts from the reputation of a product, and is a text of the content information D2 that matches the word included in the dictionary. Include the data in the text data blacklist. In this case, when the text data matching the query is included in the text data blacklist, the location information acquisition unit 114 does not transmit the location information of the entity to the web server 300 regardless of the score of the text data. Further, the web server 300 does not display the knowledge panel 220 because it does not receive the location information of the entity in response to the query included in the text data blacklist being input to the query input area 211.

As described above, in the knowledge data server 100 of the present embodiment, the score calculation unit 113 sets the target text data that does not output the entity information (location information in this example) and the entity associated with the text data. Generates the non-output content information shown (text data blacklist in this example), refers to the text data blacklist, and determines whether to output entity information in addition to the search results for the entered query. A determination unit (in this example, a location information acquisition unit 114) is provided. As a result, the knowledge data server 100 of the present embodiment can suppress the display of the knowledge panel 220 in response to a query that is not suitable for displaying the knowledge panel 220.

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.

100 ... Knowledge data server 110 ... Control unit 111 ... Selection unit 112 ... Collection unit 113 ... Score calculation unit 114 ... Location information acquisition unit 120 ... Storage unit 200 ... Terminal device 210 ... Display unit 211 ... Query input area 220 ... Knowledge panel 221 ... Entity 222 ... Image 223 ... Related information 230 ... Search result 300 ... Web server D1 ... Knowledge graphs D2, D2a ... Content information

Claims (9)

A selection unit that selects the target text data from the text data in the content information in which the entity and the text data are associated with each other.
A collection unit that collects one or more related web pages related to the target text data selected by the selection unit.
You Keru to each set of the target text data and the entity to the content information,
In the first type of related web page among one or more related web pages collected by the collecting unit, the first probability that the target text data is included as an anchor text indicating a link destination, and
Of the second probability that a link destination including the entity associated with the target text data in the content information is selected as the anchor text in the second type of related web page among the one or more related web pages. At least one and
In the first type of related web pages or the second type of related web pages, the web pages in the link destination indicated by the anchor text is the webpage entities associated with the target text data in the content information and a third probability,
Evaluation department that evaluates based on
Information processing device equipped with. The evaluation unit is the total of the number of first samples when the third probability is obtained based on the first probability and the number of second samples when the third probability is obtained based on the second probability. The value obtained by dividing the number of first samples in is multiplied by the product of the first probability and the third probability based on the first probability, and the value obtained by dividing the number of second samples by the total is defined as the second probability. Multiply the product of the third probability based on the second probability, and obtain the sum of them as the evaluation value.
The information processing apparatus according to claim 1. The first type of related web page is an encyclopedia web page that can be edited by the user.
The information processing apparatus according to claim 1 or 2. The second type of related web page is a web page that displays search results when the target text data is used as a query.
The information processing apparatus according to any one of claims 1 to 3. Further, it is provided with a determination unit that refers to the evaluation result by the evaluation unit for the input query and determines whether or not the information about the entity should be added to the search result and output.
The information processing apparatus according to any one of claims 1 to 4. The evaluation unit calculates the score of the entity corresponding to the target text data, and calculates the score.
When the score is equal to or higher than a predetermined threshold value, the determination unit outputs information on the entity to which the score is associated, and when the score is less than the predetermined threshold value, the knowledge data is associated with the entity. Decide not to output information,
The information processing apparatus according to claim 5. The determination unit refers to the target text data that is generated in advance and does not output the information of the entity and the non-output content information indicating the entity associated with the text data, and the score is equal to or higher than a predetermined threshold. Also determines that the entity included in the non-output content information is not added to the search result for the input query.
The information processing apparatus according to claim 5 or 6. The computer
Select the target text data from the text data in the content information in which the entity and the text data are associated, and select the target text data.
Collect one or more related web pages related to the selected target text data and
You Keru to each set of the target text data and the entity to the content information,
In the first type of related web page among one or more related web pages collected, the first probability that the target text data is included as an anchor text indicating a link destination, and
Of the second type of related web page among the one or more related web pages, the second probability that a link destination including the entity associated with the target text data in the content information as the anchor text is selected in the content information. , At least one and
The web page at the link destination indicated by the anchor text in the first type related web page or the second type related web page is the web page of the entity associated with the target text data in the content information . Based on 3 probabilities
Evaluate as an entity corresponding to the target text data.
Information processing method. On the computer
The target text data is selected from the text data in the content information in which the entity and the text data are associated with each other.
Have one or more related web pages related to the selected target text data collected,
You Keru to each set of the target text data and the entity to the content information,
In the first type of related web page among one or more related web pages collected, the first probability that the target text data is included as an anchor text indicating a link destination, and
Of the second type of related web page among the one or more related web pages, the second probability that a link destination including the entity associated with the target text data in the content information as the anchor text is selected in the content information. , At least one and
The web page at the link destination indicated by the anchor text in the first type related web page or the second type related web page is the web page of the entity associated with the target text data in the content information . Based on 3 probabilities
To evaluate as an entity corresponding to the target text data,
program.

Images