JP7166116B2 - Information processing device, information processing method, and program - Google Patents

Description

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

Traditionally, an electronic resource associated with a query is identified, content is selected from the electronic resource associated with the entity for display in a knowledge panel for the entity, and each content item of selected content is associated with multiple queries received. of which are selected based on the number of queries that have queried both the entity and the content represented by the content item. (For example, Patent Document 1).

JP 2016-139438 A

However, in the conventional technology, no consideration has been given to estimating a user's interests cross-sectionally among a plurality of services with different definitions of interests. Therefore, the estimated information may be limited.

The present invention has been made in consideration of such circumstances, and provides an information processing apparatus, an information processing method, and a program that can expand the estimated range of user's interests using more information. One of the purposes is to

According to one aspect of the present invention, an acquisition unit acquires a user's usage history and a user's search history regarding content information provided by a plurality of services via a network, and the content information acquired by the acquisition unit. an entity selection unit that selects entities highly related to one or more keywords included in the text data or search query of the user; an estimating unit for estimating a certain category, wherein the entity is obtained by referring to an entity representing a thing of things and a database describing relationships between the entities. .

According to one aspect of the present invention, more information can be used to expand the estimated range of user interests.

2 is a diagram showing a usage environment of the information processing apparatus 100; FIG. 4 is a diagram schematically showing processing of the information processing apparatus 100; FIG. 2 is a block diagram of the information processing apparatus 100; FIG. FIG. 4 is a diagram for explaining a knowledge base; FIG. FIG. 4 is a diagram for explaining categories; FIG. FIG. 4 is a diagram for explaining a process in which the information processing apparatus 100 collects user information; 4 is a flowchart showing an example of the flow of processing at the time of manuscript submission of the information processing apparatus 100. FIG. 4 is a flowchart showing an example of the flow of processing by the information processing device 100, the terminal device 10, and the service server 20 when recommendation information is generated and output. 10 is a flowchart showing an example of the flow of processing by the information processing device 100, the terminal device 10, and the service server 20 when generating and outputting a statistical report;

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

[Overview]
An information processing device is realized by one or more processors. An information processing device is a device that collects usage histories of a plurality of service servers and assists development across respective services.

Each of the plurality of service servers communicates with, for example, a web server that provides a web page corresponding to a request from a terminal device operated by a user, and a terminal device on which an application is started, and exchanges various information. Examples include an application server that provides content information. The service server provides, for example, news services, shopping services, auction services, matching services, financial settlement services, navigation services, etc. via the network.

The information processing device extracts one or more proper nouns from the unstructured text forming the content information provided by the service server, and sets them as keywords. Keywords are, for example, proper nouns for people, places, things, etc. that may be included in a search query entered by a user. Keywords may include aliases such as nicknames, abbreviations, common names, and self-names of the keywords in addition to the official names. Unstructured text is included in content such as news articles that include (or are related to) keywords, web pages in encyclopedias, product titles and descriptions on shopping sites, and official company information (news releases and microblogs). It is text data that can be

The information processing device estimates user's interests by accumulating entities associated with keywords for each user. In addition, the information processing device can refer to the category to which the entity belongs to enable the user's usage history of the service provided by the service server to be used in other services in the same way as the user's usage history of its own service. It makes it possible to divert it to other uses such as data collection. The categories may be shown in a hierarchical structure or a cluster structure so as to indicate the degree of similarity or association with other categories. It should be noted that the category is preferably set so that it can be used across multiple services provided by the service server 20 .

[overall structure]
FIG. 1 is a diagram showing a usage environment of an information processing apparatus 100. As shown in FIG. The information processing device 100 communicates with, for example, the service servers 20-1 to 20-N (N is a natural number) and the terminal device 10 via the network NW. The network NW includes, for example, a part or all of a WAN (Wide Area Network), a LAN (Local Area Network), the Internet, a provider device, a wireless base station, a dedicated line, and the like.

The information processing apparatus 100 stores a history of requests (for example, search requests and browsing requests) from the terminal device 10 to the service server 20 and a history of some reaction (hereinafter referred to as subsequent operation) performed by the user as a result of receiving the request. , is stored as a user usage history. In this specification, when the individual service servers 20-1 to 20-N are not distinguished, such as when explaining common matters related to the service servers 20-1 to 20-N, they are simply referred to as service servers 20. .

The terminal device 10 is a device that receives a user's input operation. The terminal device 10 is a mobile phone such as a smart phone, a tablet terminal, a personal computer, a smart speaker (AI speaker), or the like. The input operation is performed by touch operation, mouse operation, keyboard operation, voice input, or the like.

The service server 20 is a web server device that provides a web page corresponding to a request from the terminal device 10, and an application server that communicates with the terminal device 10 on which an application is started, exchanges various information, and provides content information. equipment and so on.

FIG. 2 is a diagram for explaining user usage histories collected by the information processing apparatus 100 . The information processing apparatus 100 receives, for example, requests such as search requests and browsing requests sent by the user to the service server 20 via the terminal apparatus 10, user registration information, and settings unique to services provided by the service server 20. Collect as user usage history. The information processing apparatus 100 extracts user attribute information from multiple viewpoints from the collected user usage history.

User attribute information includes, for example, demographic attributes, geographic attributes, and psychographic attributes. Demographic attributes are attribute information that can be used for labeling the social status of individual users, including, for example, age, gender, occupation, annual income, and family composition. A geographic attribute is attribute information that can be used to label an individual user according to geographical conditions, including, for example, home address, work address, and frequently visited places. Psychographic attributes are attribute information including interests, hobbies, weeks, tastes, and values. These pieces of user attribute information may be obtained from account registration information preset by the user, may be obtained from location information attached to request information transmitted from the terminal device 10, or may be obtained from the information processing device 100. may be estimated by analyzing the user usage history. The information processing apparatus 100 uses the extracted user attribute information to generate recommendation information for the user, or utilizes it for statistical analysis in selecting the content information 22 to be provided to the user, etc., so that each service server 20 provides the information. It can be used to improve services.

[Constitution]
FIG. 3 is a block diagram of the information processing device 100. As shown in FIG. The information processing apparatus 100 includes, for example, a submitted data acquisition unit 110, an analysis unit 120, an entity selection unit 130, a tag setting unit 140, a history acquisition unit 150, an estimation unit 160, an information provision unit 170, A terminal device communication unit 180 and a storage unit 190 are provided. These components of the information processing apparatus 100 except for the storage unit 190 are realized by executing a program (software) by a hardware processor such as a CPU (Central Processing Unit). Some or all of these components are hardware (circuits) such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), and GPU (Graphics Processing Unit). (including circuitry), or by cooperation of software and hardware. The program may be stored in advance in a storage device (a storage device having a non-transitory storage medium) such as the HDD or flash memory of the information processing device 100, or may be stored in a removable storage such as a DVD or CD-ROM. It is stored in a medium (non-transitory storage medium), and may be installed in the HDD or flash memory of the information processing apparatus 100 by loading the storage medium into a drive device.

The storage unit 190 is implemented by, for example, a RAM (Random Access Memory), a register, a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), or the like. The storage unit 190 stores information such as a knowledge base 192, tag information 194, usage history 196, and user attributes 198, for example.

The knowledge base 192 is a database in which information about things and information about semantic relationships between things are described as graphs. The knowledge base 192 includes, for example, abstract concepts such as “building”, “art”, and “travel”, and specific individuals belonging to or related to these concepts (“Tokyo Tower”, “Art Nouveau”, “day trip”, etc.). Includes entities that refer to "bus tours", etc.). Furthermore, the knowledge base 192 may be set with text data highly related to a specific entity. ” is set with the text data “19th century”.

The tag information 194 is information obtained by tagging the content information 22 with phrases stored in the knowledge base 192 . For example, in the news about professional baseball player A's 2000 hits, the words "sports", "professional baseball", "A player", and "2000 hits" are set as tag information.

The usage history 196 is the aforementioned user usage history. The usage history 196 includes, for example, browsing history, click history, display history, service purchase history, keywords of content information provided by the service server 20, and the like. The usage history 196 is acquired from the service server 20 by periodical batch processing or the like.

The user attributes 198 are account registration information preset by the user (eg, user ID, gender, address, occupation). The user attribute 198 may include a result estimated to be included in the user attribute information obtained by analyzing the usage history 196 (hereinafter referred to as estimated user attribute). "Sports" is set in the estimated user attribute information of the user who browses the website.

The submitted data acquisition unit 110 acquires the content information 22 of the service server 20 . This content information 22 may have been published to the user by the service server 20, or may have not been published to the user. The submitted data acquisition unit 110 outputs the unstructured text of the content information 22 to the analysis unit 120 .

The analysis unit 120 analyzes the unstructured text output by the submitted manuscript data acquisition unit 110 and sets one or more keywords. The analysis unit 120, for example, extracts words and phrases considered to be proper nouns based on the context within the unstructured text, and sets one or more keywords from the extraction results. Also, the analysis unit 120 may search for text data recorded in the knowledge base 192 to extract words that are considered to be proper nouns.

For example, if the unstructured text is a Japanese sentence, the analysis unit 120 performs morphological analysis, divides it into words, and extracts proper nouns using a technique such as W2V (Word2Vec). Other web crawling techniques may be used for the extraction of proper nouns by the analysis unit 120 . The analysis unit 120 outputs analysis results to the entity selection unit 130 .

The entity selection unit 130 scrutinizes the keywords included in the analysis results output by the analysis unit 120, and selects one or more entities highly related to the keywords. For example, when the analysis result of the analysis unit 120 includes “Japan Radio Tower”, the entity selection unit 130 selects the entity “Tokyo Tower” associated with the text data “Japan Radio Tower” in the knowledge base 192 . Note that if there is an entity with the same name as the text data of the knowledge base 192, the entity selection unit 130 may select that entity (for example, the entity "Japan Radio Tower" in the above example). The entity selection section 130 outputs the selected entity to the tag setting section 140 .

The tag setting unit 140 causes the tag information 194 to store information that associates the entity selected by the entity selection unit 130 with the content information 22 .

The history acquisition unit 150 acquires the user usage history associated with the user of interest from the usage history 196 . The history acquisition unit 150 may acquire the user usage log 24 from the service server 20 and use it as the user usage history. The user usage log 24 includes the user's usage history and the user's search history regarding content information provided via the network by a plurality of services provided by each of the service servers 20, and is information identifying the user. (For example, user ID, request type, displayed content information, presence/absence of subsequent operation). The history acquisition unit 150 outputs the acquired user usage history to the estimation unit 160 . Note that the history acquisition unit 150 is an example of an “acquisition unit”.

Based on the entity selected by the entity selection unit 130 and the user usage history acquired by the history acquisition unit 150, the estimation unit 160 determines the category associated with the entity that the user is interested in as the user's Presumed to be a category of interest. A method of estimating the category will be described later. The estimation unit 160 stores the estimated user's interest category as a user attribute 198 .

For example, if the user's subsequent operation is included in the user usage history, the estimation unit 160 estimates that the user is likely to be interested in the category associated with the selected entity. A flag, rank, or the like indicating that there is a high possibility is set and stored as a user attribute 198 .

The information providing unit 170 acquires search conditions related to the user's interests from the service server 20, searches categories accumulated in the user attributes 198 based on the search conditions, and returns the obtained search results to the service server 20. do.

For example, the information providing unit 170 may use search conditions related to the user's interests included in a request to view predetermined content information (for example, a top page or a page customized for each user) of the service server 20 transmitted by the terminal device 10. is received, and a search result extracted with the search condition for the accumulated user attribute 198 is generated. The information providing unit 170 extracts content information to be recommended to the user based on the categories estimated by the estimation unit 160 and accumulated in the user attribute 198 and estimated to be of interest to the user.

The information providing unit 170 searches for content information in which tag information highly relevant to the category associated with the tag information 194 is set, using, for example, the category stored in the user attribute 198 as a search key. The search results generated by the information providing unit 170 may be recommended information provided to a single user, or may be provided to a group of users presumed to have the same or similar interests. It may be recommendation information that In addition, the information providing unit 170 may generate recommendation information in a ranking format with some or all of the users as aggregation targets, as recommendation information to be provided to users whose interests are not estimated. The information providing unit 170 outputs the generated search result to the terminal device communication unit 180 .

In addition, the information providing unit 170 receives recommendation information from the user of the terminal device 10 (for example, the service provider of the service server 20) who has the authority to browse the recommendation information of other users. The recommended information that satisfies the conditions specified by the user of the terminal device 10 who received the output request may be output in a report format or the like. The report-format recommendation information output by the information providing unit 170 is hereinafter referred to as a statistical report.

The conditions specified by the user of the terminal device 10 are the demographic attributes stored in the user attributes 198 stored in a manner that can be compared with the stored database of categories that are presumed to be of interest to the user. or should be included in the geographic attributes. For example, the service provider of the service server 20 may provide information such as the number of users who are interested in the category "Yokohama baseball team", statistical information by gender and age of the users, and other categories in which those users tend to be interested. can be obtained from the information processing device 100.

The terminal device communication unit 180 returns the search results output by the information providing unit 170 corresponding to the user to the terminal device 10 . For example, when the terminal device 10 displays the specific content of the service server 20, the terminal device communication unit 180 receives a recommendation information output request, and returns the recommendation information associated with the user of the terminal device 10 as a response. do.

[Category Estimation]
A category estimation method by the estimation unit 160 will be described below. FIG. 4 is a diagram for explaining entities stored in the knowledge base 192. As shown in FIG.

The things described in the knowledge base 192 are defined by an ontology. An ontology defines classes and properties of things, and is a collection of constraints between classes and properties.

For example, a beaked, egg-laying vertebrate with winged forelimbs would be classified as a "bird". In addition, in the class "bird", an entity having the property of being unable to fly is classified into lower classes such as "penguin" and "ostrich". In this way, the system of classes may have a hierarchical structure having a relationship between upper and lower levels. Properties of higher classes are inherited by lower classes. In the above example, the property of the class ``bird'', ``a vertebrate that has a beak, lays eggs, and has winged forelimbs'', is similar to the property of the lower classes ``penguin'' and ``ostrich''. will also be included. As for the class name for identifying the class, for example, if the class is "bird", the class name may be represented by the character string "bird". Note that the class name does not necessarily have to represent a meaning. For example, even for the class "bird", a character string indicating simple identification information such as "information 1" or "C1" may be assigned. . The above-mentioned entities, that is, things, belong to any class included in the class system defined by the ontology.

A property is an attribute that describes the properties and characteristics of things and the relationships between classes. For example, a property may be an attribute that indicates the property of "having ~ as a component of the body", or the property of "inhabiting ~", or "a class is a superclass and a class is a subclass". It may be an attribute that indicates a higher-lower relationship between classes such as "is a class". A property name for identifying a property may or may not have a meaning, like the class name described above.

The knowledge base 192 is a directed graph in which the above classes are represented as nodes and the above properties are represented as labeled and directed edges. With such a graph structure, it is possible to determine information about things by nodes and relationships between things by edges.

The above properties can also be transferred to describe relationships between entities. For example, the knowledge base 192 shown in FIG. 4 defines a relationship between entities such that entity E1 belongs to the same class as entities E2 and E3. The entity E1 and the entity E2 are represented by the property "affiliation team", and the entity E1 and the entity E3 are represented by the property "occupation".

FIG. 5 is a diagram for explaining categories. An example in which categories are shown in a hierarchical structure will be described below. The administrator of the information processing apparatus 100 associates categories with entities based on a category system common among a plurality of services, and uses the categories to estimate the user's interests. The administrator of the information processing apparatus 100 sets, for example, the entity E2 “Yokohama team” in FIG. 4 as being associated in advance with the category C111 “Yokohama team” in FIG.

Therefore, when the entity E1 "A player" is set as a keyword for the analysis unit 120, the entity selection unit 130 selects the category C111 "Yokohama team", or selects the category C1 " sports” or category C11 “professional baseball”. The estimation unit 160 stores the selected category in the user attribute 198 as an estimated user attribute.

In addition, the estimation unit 160 may estimate and assign the strength of the user's interest in the category to the estimated user attribute. The estimating unit 160, for example, refers to the time series of usage histories, weights categories related to new usage histories so that higher points are set, and expresses the strength of interest in numerical values or ranks. do.

[Recommendation target extraction]
A method for extracting content information to be recommended by the information providing unit 170 will be described below. FIG. 6 is a diagram for explaining the relationship between usage history 196 and user attributes 198. As shown in FIG.

The upper diagram in FIG. 6 is a diagram showing an example of the usage history 196 . In the usage history 196, for example, a service that provides the content information 22 associated with the user usage history, the type of history, a keyword associated with the content information 22, presence/absence of subsequent operations, and the like are set. For example, the information providing unit 170 extracts a category associated with the content information to be recommended from the usage history 196 . When the estimation unit 160 estimates the level of interest of the user in the category, the information providing unit 170 may change the category to be extracted according to the level of interest.

The lower diagram in FIG. 6 is a diagram showing an example of user attributes 198 . When selecting a category, the information providing unit 170 may divert the service information from the usage history 196 as it is, or divert it to a service different from the usage history 196 if the category can be diverted to other services. may For example, the information providing unit 170 may use the category "professional baseball" associated with the service "news" as the recommendation information as it is, or may use the category "Yokohama" associated with the service "news" as the content information of the same category. It may be used as recommendation information by diverting to the service "Gourmet".

The information providing unit 170 may extract the above-described recommendation target for each user, and may select groups of users with similar psychographic attributes (i.e., categories with the same or similar interests are stored as the user attributes 198). The above-described recommendation target extraction may be performed for each user group. Further, the information providing unit 170 may use the extraction results of recommendation targets of a plurality of user groups as recommendation information for users without attribute information.

[Upload processing flow]
FIG. 7 is a flow chart showing an example of the processing flow of the information processing apparatus 100 when the content information 22 is submitted to the service server 20. As shown in FIG.

First, the submitted data acquisition unit 110 acquires the content information 22 (S100). Next, the analysis unit 120 sets keywords (S102). Next, the entity selection unit 130 selects an entity (S104). Next, the tag setting unit 140 stores, as tag information 194, information that associates the entity selected by the entity selection unit 130 with the content information 22 (S106). This completes the description of the processing of this flowchart.

[Recommendation extraction processing flow]
FIG. 8 is a flow chart showing an example of the processing flow of the information processing device 100, the terminal device 10, and the service server 20 when generating and outputting recommendation information. The processing from S200 to S204 is performed periodically, and the processing from S206 to S212 is performed when the user uses the service server 20. FIG.

First, the history acquisition unit 150 acquires user usage history (S200). Next, the estimation unit 160 estimates the user's interest category based on the user's usage history (S202). Next, the information providing unit 170 extracts recommendation information based on the user's interest category (S204).

The service server 20 receives the target user's service display request from the terminal device 10 (S206). Next, the terminal device communication unit 180 receives a recommendation information output request from the service server 20 (S208), and transmits the search result of recommendation information as a response (S210). Next, the service server 20 generates display service information for the target user based on the received search results (S212), and transmits the generated display service information to the terminal device 10 (S214). End the explanation of the processing. Such processing can expand the estimated range of the user's interests.

[Provision flow of statistical results based on interests]
FIG. 9 is a flow chart showing an example of the flow of processing by the information processing device 100, the terminal device 10, and the service server 20 when generating and outputting a statistical report.

First, the service provider of the service server 20 transmits a statistical report output request specifying search conditions to the information processing apparatus 100 via the terminal device 10 (S300). Next, the terminal device communication unit 180 receives the output request (S302). Next, the information providing unit 170 extracts users who meet the search conditions (S304), and generates a statistical report based on the recommendation information associated with the users (S306). The terminal device communication unit 180 transmits the statistical report to the terminal device 10 of the service provider (S308). This completes the description of the processing of this flowchart.

According to the embodiment described above, the history acquisition unit 150 acquires the user's usage history and the user's search history regarding the content information 22 provided via the network by a plurality of services, and the history acquisition unit 150 acquires an entity selection unit 130 that selects an entity highly related to one or more keywords included in the text data of the content information 22 or the search query from the entities stored in the knowledge base 192; By providing an estimating unit 160 for estimating categories associated with entities in which the user is interested, it is possible to expand the estimated range of the user's interests using more information. .

As described above, the mode for carrying out the present invention has been described using the embodiments, but the present invention is not limited to such embodiments at all, and various modifications and replacements can be made without departing from the scope of the present invention. can be added.

REFERENCE SIGNS LIST 10 terminal device 20 service server 22 content information 24 user usage log 100 information processing device 110 manuscript data acquisition unit 120 analysis unit 130 entity selection unit 140 tag setting Unit 150 History acquisition unit 160 Estimation unit 170 Information provision unit 180 Terminal device communication unit 192 Knowledge base 194 Tag information 196 Usage history 198 User attribute

Claims (7)

an acquisition unit that acquires a user's usage history regarding content information provided via a network by a plurality of services and the user's search history;
an entity selection unit that selects an entity highly related to one or more keywords included in the text data or search query of the content information acquired by the acquisition unit;
an estimation unit for estimating the category associated with the entity selected by the entity selection unit as the category of interest of the user;
The entity is obtained by referring to a database describing entities representing things and relationships between the entities,
Categories of interests of the user are stored in a database, search conditions related to interests specifying user attributes are acquired, and the number or percentage of users who are interested in a specific category for each user attribute is stored in the database. Equipped with an information providing unit that returns as a search result of
Information processing equipment. The estimation unit estimates a category of interest of the user based on a category system common among the plurality of services.
The information processing device according to claim 1 . The information providing unit further acquires, as the search condition, user attribute information of the user stored in advance in a manner that can be compared with the database,
The information processing apparatus according to claim 1 or 2. The estimation unit estimates that there is a high possibility that the category associated with the entity is of interest when the user's subsequent operation is included in the usage history.
The information processing apparatus according to any one of claims 1 to 3. further comprising an information providing unit that extracts the content information to be recommended to the user based on the category estimated by the estimation unit;
The information processing apparatus according to any one of claims 1 to 4. the computer
Acquiring a user's usage history and the user's search history regarding content information provided by a plurality of services via a network,
selecting an entity highly relevant to one or more keywords included in the text data or search query of the acquired content information;
estimating the category associated with the selected entity as the category of interest of the user;
The entity is obtained by referring to a database describing entities representing things and relationships between the entities,
The computer further
accumulating categories of interest of the user in a database;
Acquire search conditions related to interests that specify user attributes,
Return the number or percentage of people who are interested in a specific category for each user attribute as a database search result,
Information processing methods. to the computer,
Acquiring a user's usage history and the user's search history regarding content information provided by a plurality of services via a network;
selecting an entity highly relevant to one or more keywords included in the text data or search query of the acquired content information;
presuming that the category associated with the selected entity is the category of interest of the user;
The entity is obtained by referring to a database describing entities representing things and relationships between the entities,
The computer further
accumulating categories of interest of the user in a database;
Acquire search conditions related to interests that specify user attributes,
Return the number or percentage of people who are interested in a specific category for each user attribute as a search result of the database,
program.

Images