Information processing device, information processing method, and information processing program

Abstract

To determine content items recommended for one or more users constituting a family.SOLUTION: An information processing apparatus is configured to: create a relationship graph indicating a social relationship among multiple users based on user characteristics of the multiple users; using the relationship graph to predict family users including, among the multiple users, a registered user registered with a content distribution service and co-residing users living together with the registered user; identify which of the family users has viewed each of multiple content items provided to the registered user by the content distribution service; and, based on results of the identification, determine recommendation content items recommended for the family users by means of a trained model.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program, and in particular to a technique for recommending content to a user. [Background technology]

[0002] In recent years, content distribution services that provide video content over the Internet have become widespread. In content distribution services, it is important to recommend media content, including video content such as movies and animations, and advertising content, in order to further improve the use of the service. Patent Document 1 discloses a method for displaying content recommendations selected based on knowledge of a user's interests and activities (e.g., a user's search queries, search results, viewing history, purchase history, and a user's physical activity). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2022-527229 Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, a content distribution service provides content to users who have registered with the service (registered users). By using the technology disclosed in Patent Document 1, it becomes possible to recommend content suitable for the registered users.

[0005] On the other hand, when a family consisting of multiple users including the registered user shares a digital device (e.g., a smart TV or a tablet device) for viewing content, the viewed content may be from a user other than the registered user. In other words, it is basically not known which family member viewed the content viewed on the shared digital device. Conventionally, in such a case, there was no mechanism for determining the content to be recommended to the whole family or to each user, taking into account the characteristics of users other than the registered user in the family.

[0006] The present invention has been made in consideration of the above-mentioned problems, and has an object to provide a mechanism for determining content to be recommended to one or more users who make up a family. [Means for solving the problem]

[0007] In order to solve the above problem, one aspect of an information processing device according to the present invention includes a graph creation means for creating a relationship graph showing social relationships between a plurality of users based on the user characteristics of the plurality of users; a prediction means for predicting family users among the plurality of users using the relationship graph, including registered users who are registered with a content distribution service and co-resident users who live with the registered users; an identification means for identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; and a decision means for determining recommended content to be recommended to the family user using a learning model based on the results of the identification.

[0008] In order to solve the above problem, one aspect of an information processing method according to the present invention includes a graph creation step of creating a relationship graph showing social relationships between a plurality of users based on user characteristics of the plurality of users; a prediction step of predicting family users among the plurality of users using the relationship graph, including registered users who are registered with a content distribution service and co-resident users who live with the registered users; an identification step of identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; and a determination step of determining recommended content to be recommended to the family user using a learning model based on the results of the identification.

[0009] In order to solve the above problem, one aspect of the information processing program of the present invention is an information processing program for causing a computer to execute information processing, the program causing the computer to execute processes including: a graph creation process for creating a relationship graph showing social relationships between a plurality of users based on the user characteristics of the plurality of users; a prediction process for predicting family users among the plurality of users, including registered users who are registered with a content distribution service and co-resident users who live with the registered users, using the relationship graph; an identification process for identifying which of the family users viewed each of a plurality of contents provided to the registered user by the content distribution service; and a decision process for determining recommended content to be recommended to the family user using a learning model based on the results of the identification. Effect of the Invention

[0010] According to the present invention, a mechanism is provided for determining content to be recommended to one or more users who make up a family. The above-mentioned objects, aspects, and advantages of the present invention, as well as objects, aspects, and advantages of the present invention not described above, will be understood by those skilled in the art from the following detailed description of the invention by referring to the accompanying drawings and the claims. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 shows an example of the configuration of an information processing system according to an embodiment. [Diagram 2] FIG. 2 is a schematic diagram illustrating a use case assumed in the embodiment. [Diagram 3] FIG. 3 illustrates an example of a functional configuration of an information processing device according to an embodiment. [Figure 4] FIG. 4 shows a flowchart of the social graph creation process. [Figure 5A] FIG. 5A is a diagram illustrating an explicit link in a social graph. [Figure 5B] FIG. 5B is a diagram illustrating implicit links in a social graph. [Figure 6A] FIG. 6A is a diagram for explaining the process of inferring the relationship between links. [Figure 6B] FIG. 6B shows a flow chart of an example process for grouping pairs into clusters. [Figure 7A] FIG. 7A shows a conceptual diagram of a closeness score based on the closeness of a relationship for a pair of users. [Figure 7B] FIG. 7B shows the schematic architecture of the intimacy score prediction model. [Figure 7C] FIG. 7C shows a conceptual diagram of a social graph. [Figure 8A] FIG. 8A is a diagram for explaining a learning procedure of the viewing score prediction model. [Figure 8B] FIG. 8B is a diagram for explaining the learning procedure of the viewing score prediction model. [Figure 8C] FIG. 8C is a diagram for explaining the learning procedure of the viewing score prediction model. [Figure 9A]FIG. 9A is a diagram for explaining a procedure for determining recommended content. [Figure 9B] FIG. 9B is a diagram for explaining a procedure for determining recommended content. [Figure 10] FIG. 10 is a flowchart showing the overall process executed by the information processing device according to the embodiment. [Figure 11] FIG. 11 shows an example of the hardware configuration of an information processing device and a communication device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, with reference to the attached drawings, an embodiment for carrying out the present invention will be described in detail. Among the components disclosed below, those having the same functions are given the same reference numerals, and their description will be omitted. Note that the embodiment disclosed below is an example of a means for realizing the present invention, and should be appropriately modified or changed depending on the configuration of the device to which the present invention is applied and various conditions, and the present invention is not limited to the following embodiment. Furthermore, not all of the combinations of features described in the present embodiment are necessarily essential to the solution of the present invention.

[0013] [Example of information processing system configuration] Fig. 1 shows an example of the configuration of an information processing system according to this embodiment. As an example, as shown in Fig. 1, this information processing system includes an information processing device 10 and a number of communication devices 11-1 to 11-N (N>1) used by a number of arbitrary users 1 to N. In the following description, the communication devices 11-1 to 11-N may be collectively referred to as communication device 11 unless otherwise specified.

[0014] The communication device 11 is a device having a display unit (display surface), such as a smartphone, a tablet terminal, or a smart TV. The communication device 11 is configured to be able to communicate with the information processing device 10 via a public network such as LTE (Long Term Evolution) or a wireless communication network such as a wireless LAN (Local Area Network). One user may use multiple communication devices 11, and in this case, each communication device 11 is configured to be able to communicate with the information processing device 10. When the communication device 11 is a device having a GUI (Graphic User Interface) equipped on a display unit such as a liquid crystal display, such as a smartphone or a tablet terminal, the user can perform various operations using the GUI. The operations include various operations on content such as images displayed on a screen, such as a tap operation, a slide operation, and a scroll operation using a finger or a stylus. The communication device 11 is not limited to the device shown in Fig. 1, and may be a device such as a desktop PC (Personal Computer) or a notebook PC. In this case, each user may operate the device using an input device such as a mouse or a keyboard. The communication device 11 may also be provided with a separate display screen.

[0015] The communication device 11 can use web services (internet-related services) provided by the information processing device 10 or from another device (not shown) via the information processing device 10. The web service may include a content distribution service (streaming service). The content distribution service is a service that provides video content such as movies, animations, and dramas, and may also provide advertising content (videos or still images). In the present disclosure, the video content is also referred to as culture content in order to distinguish between video content such as movies, animations, and dramas and advertising content. The web services may also include online malls and online supermarkets, as well as services related to communications, finance, real estate, sports, and travel.

[0016] These web services can be used by users, for example, by registering for use with an API (Application Programming Interface) provided by the information processing device 10 or another device. The web services include any membership service, and there is no restriction on the type of service as long as they are services that can observe or manage interactions (actions) between entities. When registering for use with a web service, each user is associated with a user ID that identifies the user.

[0017] A user transmits information about the user and the communication device 11 used by the user to the information processing device 10 through the use of a web service. For example, the communication device 11 can transmit information such as the IP (Internet Protocol) address of the communication device 11, the user's address, and the user's name (first and last name) to the information processing device 10. In addition, a user who uses a content distribution service views content on the communication device 11, and information on the history of the viewed content (viewing history) is transmitted to the information processing device 10.

[0018] Here, one use case assumed in this embodiment will be described with reference to Fig. 2. Fig. 2 is a schematic diagram for explaining a use case assumed in this embodiment. As described above, in this embodiment, a web service provided from the information processing device 10 or from another device (not shown) via the information processing device 10 can include a content distribution service. In this disclosure, a user who has registered to use a content distribution service in order to use the service is referred to as a "registered user."

[0019] The registered user 20 is a user who has registered to use the content distribution service. FIG. 2 shows that family users (family members) 23 including the registered user 20 use a communication device 11 to view a series of contents 26 provided to the registered user 20 by the content distribution service. The communication device 11 may be, for example, a smart TV 251 or a tablet terminal 252. The family users 23 are members of a family (household) including the registered user 20 who resides with the registered user 20. The cohabiting users 24 are family users of the family users 23 other than the registered user 20. The cohabiting users 24 include, for example, a spouse user 21 who is the spouse of the registered user 20, and a child user 22 who is the child of the registered user 20.

[0020] The series of contents 26 are contents viewed by the communication device 11 used by the family users 23, and include contents 26-1 to 26-m (m>0). Each content is assigned a content ID for identifying the content. Each content may be culture content or advertising content. Each content may be content other than culture content or advertising content. As described later, each content is assigned metadata, which is information indicating the characteristics of the content, as auxiliary information. Each family user 23 can view a content selected from a plurality of contents provided to the communication device 11 by a predetermined operation. Information on the viewing history of the viewed content is transmitted from the communication device 11 to the information processing device 10.

[0021] In FIG. 2, the family user 23 (registered user 20, spouse user 21, and child user 22) is any one of users 1 to N shown in FIG. 1. Therefore, the information processing device 10 is configured to be able to acquire information through the use of web services by the registered user 20, spouse user 21, and child user 22. Meanwhile, the registered user 20 is registered in the content distribution service, but the spouse user 21 and the child user 22 are not registered. Therefore, the information processing device 10 recognizes the viewing history of the content acquired from the communication device 11 used by the family user 23 to view the content (viewing history by the family user 23) as the viewing history of the registered user 20. That is, the information processing device 10 cannot recognize the viewing history by the family user 23 as being explicitly distinguished from the viewing history of the registered user 20, spouse user 21, and child user 22.

[0022] The information processing device 10 according to this embodiment predicts cohabiting users 24 (spouse user 21 and child user 22 in the example of FIG. 2) who live with the registered user 20 (i.e., predicts family users 23). The information processing device 10 then predicts which of the family users 23 has viewed each piece of content provided to the registered user 20 by the content distribution service. Furthermore, the information processing device 10 determines content to be recommended to the family users 23 as a whole and / or to each family user 23 (hereinafter, recommended content) based on the prediction result.

[0023] Specifically, first, the information processing device 10 acquires various information from the communication devices 11-1 to 11-N used by the users 1 to N, and creates a social graph (relationship graph) based on the information. The social graph is a graph network showing social relationships between multiple users. Second, the information processing device 10 predicts and detects cohabiting users 24 based on the social graph. Third, the information processing device 10 predicts (identifies) which of the family users 23 has viewed each of the contents provided to the registered user 20 by the content distribution service. Then, based on the prediction result (identification result), the information processing device 10 trains a viewing score prediction model 113 for each family user 23, and generates a viewing score prediction model 113 for all the family users 23. The information processing device 10 uses the viewing score prediction model 113 to determine recommended content for each family user 23 and / or all the family users 23.

[0024] [Functional configuration of information processing device 10] FIG. 3 shows an example of a functional configuration of the information processing device 10 according to the present embodiment. The information processing device 10 shown in Fig. 3 includes a user feature acquisition unit 101, a viewing history acquisition unit 102, a social graph creation unit 103, a family user prediction unit 104, a viewing user prediction unit 105, a learning unit 106, a recommended content determination unit 107, an output unit 108, a learning model storage unit 110, and a data storage unit 120. The learning model storage unit 110 stores a user feature prediction model 111, an intimacy score prediction model 112, and a viewing score prediction model 113. The various learning models will be described later. The data storage unit 120 stores a user feature 121 and a viewing history 122. In describing Fig. 3, Figs. 1 and 2 will be referred to.

[0025] The user feature acquisition unit 101 acquires factual features (factual information) about the communication devices and users from each of the communication devices 11-1 to 11-N as user features. The user features are factual features (information) that are actually or objectively obtained from the communication devices and users. The user feature acquisition unit 101 can acquire the user features directly from the communication device 11, for example. The user feature acquisition unit 101 can also acquire the user features as information registered in a predetermined web service by the user of the communication device 11. The user features are associated with a user ID that identifies the user.

[0026] The user characteristics include the IP address of the communication device 11, the user's address, the user's name, the credit card number held by the user, the user's demographic information (demographic user attributes such as gender, age, area of ​​residence, occupation, family composition, etc.), etc. The user characteristics may include a registration number and a registered name when using a web service, a call history, a delivery address other than the user's address when using a certain web service, a usage status when using a web service, a purchase history, a usage history, a search history, and information about points that can be accumulated by using a service. The user characteristics may also include characteristics related to the user's home time, such as the day of the week and the time the user is at home. The characteristics related to the home time may be obtained, for example, from a delivery date and time specified by the user or an actual delivery date and time when the user purchases an item using a web service. The user characteristics may also include characteristics related to preferences, such as whether the user smokes tobacco, drinks alcohol, etc. The characteristics related to preferences may be acquired based on items purchased by the user using a web service, for example. Thus, the user characteristics may include any information, including information related to the communication device 11 or the user himself / herself, and information related to the use of a given service via communication.

[0027] The user feature acquisition unit 101 is also configured to apply the acquired user features to the trained user feature prediction model 111 to acquire user features (estimated user features (attributes)) estimated for the user features. The user feature prediction model 111 is configured to receive the user features (i.e., fact features) of a target user as input, and output the probability (correspondence probability) that each of the multiple user features corresponds (matches) the target user. The user feature acquisition unit 101 ultimately determines the estimated user features of the target user from the correspondence probability.

[0028] For example, the user feature acquisition unit 101 inputs the demographic information of the target user and data indicating a purchase history and preferences in past web services of the target user as the user features of the target user to the user feature prediction model 111. The user feature prediction model 111 outputs, as the corresponding probability, the probability for each of a plurality of items that the target user is estimated to purchase, a plurality of preferences that the target user may have, and the behavioral tendency of the target user. Then, the user feature acquisition unit 101 can acquire the items, preferences, and behavioral tendencies having a probability equal to or greater than a predetermined value as the estimated user features of the target user. The behavioral tendencies include, for example, characteristics related to a life stage, such as having children, and characteristics related to a lifestyle, such as working from home.

[0029] The user feature acquisition unit 101 stores the acquired and estimated user features in the data storage unit 120 as user features 121. The user features are stored in association with a user ID.

[0030] The viewing history acquisition unit 102 acquires content viewing history information from the communication devices 11 of registered users who are registered with the content distribution service among users 1 to N. The viewing history information acquired by the viewing history acquisition unit 102 corresponds to the viewing history information of the registered users. With reference to FIG. 2, the viewing history acquisition unit 102 acquires a history of content viewed by any of the family users 23 as the viewing history of the registered user 20. The viewing history information includes information on the content ID, metadata, and viewing date and time (viewing time) of each content viewed.

[0031] Metadata (attached information) is information given to each content and indicates the characteristics of the content. For example, in the case of cultural content such as movies, animations, and dramas, the metadata may include information such as the genre, the target age group of the content, the title, the cast (actors, etc.), the director, the content time, the series number, and the plot. In the case of advertising content, the metadata may include information such as the advertiser, the target item (product or service), and the genre. The viewing history acquisition unit 102 stores the acquired viewing history information in the data storage unit 120 as the viewing history 122. The viewing history 122 is stored in association with the user ID. Referring to FIG. 2, the viewing history acquired from the registered user 20 is stored in association with the user ID of the registered user 20.

[0032] The social graph creation unit 103 creates a social graph (relationship graph) by using the various user features acquired by the user feature acquisition unit 101. The procedure for creating a social graph will be described later.

[0033] The family user prediction unit 104 predicts (searches for) family users including registered users of the content distribution service. First, the family user prediction unit 104 acquires a user ID of a registered user who is one of users 1 to N, and identifies the registered user. The user ID of the registered user may be input by an operator via an input unit (input unit 1105 in FIG. 11) or via a communication unit (communication I / F 1107 in FIG. 11). Alternatively, the user ID of the registered user may be set by an arbitrary program stored in a storage unit (ROM 1102 or RAM 1103 in FIG. 11). The family user prediction unit 104 acquires the user characteristics of the registered user from the user characteristics 121 of the data storage unit 120 based on the user ID of the registered user.

[0034] Then, the family user prediction unit 104 predicts family users including the registered user based on the acquired user characteristics of the registered user and the social graph created by the social graph creation unit 103. With reference to FIG. 2, the family user prediction unit 104 predicts cohabiting users 24 based on the user characteristics and social graph of the registered user 20. Note that since there are cases where the registered user is a single person and has no cohabiting users (number of family users=1), the family user prediction unit 104 is configured to be able to determine whether or not there are any cohabiting users for the registered user. The procedure for predicting family users based on the social graph will be described later.

[0035] Based on the viewing history of the registered user included in the viewing history 122 and the family users predicted by the family user prediction unit 104, the viewing user prediction unit 105 predicts (identifies) one or more users (hereinafter, viewing users) who have viewed each of the one or more contents included in the viewing history. That is, the viewing user prediction unit 105 predicts which one or more family users have viewed each of the one or more contents included in the viewing history of the registered user. The prediction corresponds to the association (alignment) of each of the one or more contents with each of the family users. Referring to FIG. 2, the viewing user prediction unit 105 predicts the viewing users of each of the contents 26. That is, the viewing user prediction unit 105 predicts (identifies) which of the registered user 20, the spouse user 21, and the child user 22 has viewed each of the contents 26.

[0036] In this embodiment, the viewing user prediction unit 105 makes the prediction by referring to metadata and / or viewing date and time of one or more contents included in the viewing history of the registered user, and the user characteristics of the family user 23 acquired from the user characteristics 121. For example, the viewing user prediction unit 105 may make the prediction by using common characteristics between the metadata and / or viewing date and time, and the user characteristics. With reference to FIG. 2, it is assumed that the metadata of the content 26-1 includes the genre=anime and the target age group of the content=3 to 10 years old, the viewing date and time is a weekday evening, and the age included in the user characteristics of the child user 22 is 8 years old. In this case, since the target age of the content included in the metadata of the content 26-1 and the age characteristics of the child user 22 are common, the viewing user prediction unit 105 can predict that the viewing user of the content 26-1 is the child user 22.

[0037] Alternatively or in addition to this, the viewing user prediction unit 105 may make the prediction based on a rule. The rule is a rule that associates metadata and / or viewing date and time of the content with user characteristics. An example of the rule is a rule that indicates that a viewing user of a content whose metadata includes genre=anime and whose viewing date and time is in the evening on a weekday is a user whose user characteristics include an age of 6 to 12 years old and / or being at home in the evening on a weekday. When the metadata of the content 26-2 includes genre=anime and whose viewing date and time is in the evening on a weekday, the viewing user prediction unit 105 can predict that the viewing user of the content 26-2 is a child user 22 (8 years old) based on the exemplary rule. Furthermore, when the user characteristics 121 of the spouse user 21 include being at home in the evening on a weekday, the viewing user prediction unit 105 can predict that the viewing user of the content 26-2 further includes the spouse user 21 based on the exemplary rule.

[0038] Moreover, the rule may be a rule that associates metadata and / or viewing date and time of the content with behavioral tendencies included in the user characteristics. For example, a rule is set that associates content having metadata of a long content time (e.g., more than two hours) with a user whose behavioral tendencies include working from home. The exemplary rule is based on a prediction (e.g., empirical rule) that a user who works from home is more likely to watch content with a long content time because he or she saves commuting time and spends more time at home. When the metadata of the content 26-3 includes a long content time and the behavioral tendencies of the registered user 20 include working from home, the viewing user prediction unit 105 can predict that the viewing user of the content 26-3 is the registered user 20. This allows the viewing user prediction unit 105 to predict the viewing user of each content according to the behavioral tendencies of each family user.

[0039] The rule may be created by, for example, empirical rules or machine learning. The rule may be input by an operator via an input unit (input unit 1105 in FIG. 11) or via a communication unit (communication I / F 1107 in FIG. 11). Alternatively, the rule may be set in advance in the system or may be set by an arbitrary program stored in a storage unit (ROM 1102 or RAM 1103 in FIG. 11).

[0040] Through such prediction processing, the viewing user prediction unit 105 associates each of the contents included in the viewing history 122 with one or more viewing users. Furthermore, the viewing user prediction unit 105 generates (prepares) combination data of each of the contents included in the viewing history 122 with one or more viewing users from the association. The combination data includes the content ID and metadata of the content, and user characteristics of the viewing data associated with the content. The combination data may also include information on the viewing date and time of the content.

[0041] The learning unit 106 trains the viewing score prediction model 113 using the combination data generated by the viewing user prediction unit 105. The viewing score prediction model 113 is a learning model for machine learning that predicts the viewing score of a content by inputting the combination data of metadata of the content and the user characteristics of the user. The viewing score indicates the probability (possibility) that the input content is viewed by a user having the input user characteristics, and is expressed as a numerical value of 0 to 1 (maximum probability=1), for example. The viewing score indicates not only the probability of viewing, but also the probability of the content being selected by any action, such as the probability of the content being selected by a click action on a screen on which the content is displayed. A learning procedure of the viewing score prediction model 113 by the learning unit 106 will be described later.

[0042] The recommendation content determination unit 107 determines the recommendation content by using the viewing score prediction model 113 learned by the learning unit 106. Referring to FIG. 2, the recommendation content determination unit 107 determines the recommendation content for the family user 23. The procedure for determining the recommendation content will be described later.

[0043] The output unit 108 outputs the recommendation content determined by the recommendation content determination unit 107. For example, the output unit 108 generates information regarding the recommendation content and outputs (displays) the information to the display unit of the communication device 11 used by the registered user. The output unit 108 may generate and output information regarding the recommendation content. The output may include any form of output such as output by display or output by voice.

[0044] [Procedure for creating a social graph] Next, the procedure for creating a social graph according to the present embodiment will be described. In the following description, users A to E are users referred to for the purpose of explanation and may be users of the communication device 11. The social graph is composed of connections of a plurality of entities (corresponding to the user nodes surrounded by circles in the examples of FIGS. 5A and 5B). In the following description, the user node is simply referred to as a user. FIG. 4 shows a flowchart of the social graph creation process executed by the social graph creation unit 103 according to the present embodiment. Each step of the process in FIG. 4 will be described below.

[0045] <S41: Creation of explicit and implicit links in the social graph> In S41, the social graph creation unit 103 predicts and creates explicit and implicit links between a plurality of users. The process of creating links in a social graph will be described with reference to Figures 5A and 5B. Figure 5A is a diagram for explaining an explicit link in a social graph, and Figure 5B is a diagram for explaining an implicit link in a social graph. An explicit link in a social graph is a link created based on an explicit common user characteristic between two users (a user pair). An implicit link in a social graph is a link created as an indirect relationship by utilizing an explicit link that has already been created, even though the existence of an explicit common user characteristic between a user pair is unclear.

[0046] Fig. 5A shows an example of creating an explicit link in a social graph using the IP addresses of users' communication devices as a common user feature. Fig. 5A shows an example in which there are an online mall 51, a golf course reservation service 52, a travel-related reservation service 53, and a card management system 54 as web services available to users A to C. Fig. 5A shows these four web services, but the number of web services is not limited to a specific number.

[0047] The online mall 51 is a shopping mall available online (using the Internet). The online mall 51 can provide a wide variety of products and services, such as fashion, books, food, concert tickets, and real estate. The golf course reservation service 52 is operated on a website that provides online services related to golf courses, and can provide, for example, golf course search and reservation and lesson information. The travel-related reservation service 53 is operated as a website that provides various travel services available online. The travel-related reservation service 53 can provide, for example, hotel and travel tour reservations, airline and rental car reservations, tourist information, and information about hotels and the surrounding areas of hotels. The card management system 54 is operated on a website that provides services related to credit cards issued and managed by a specific card management company. The card management system 54 may provide services in association with at least one of an online mall 51, a golf course reservation service 52, and a travel-related reservation service 53.

[0048] 5A, users A to C each use the same IP address (=198.45.66.xx) to access an online mall 51, a golf course reservation service 52, and a travel-related reservation service 53. IP address information can be acquired by the user characteristic acquisition unit 101. In such a case, as shown in link state 55, social graph creation unit 103 creates explicit links (solid line links) between users A to C in the social graph with the same IP address characteristic.

[0049] In addition to FIG. 5A, explicit links in the social graph can be created using common user features such as the user's address feature and the feature of the credit card number used by the user.

[0050] FIG. 5B shows an example of creating an implicit link between users in a social graph. In the example of FIG. 5B, for user A, users C, D, and E are connected by explicit links, and for user B, users C, D, and E are connected by explicit links. Such link features (features indicating the relationship between links) are embedded into a common feature space, and links inferred to implicitly establish a relationship between each user (each node) are created (established) as implicit links. In the example of FIG. 5B, although user A and user B are not connected by an explicit link, an implicit link (dotted link) is created as a result of being inferred to have a relationship in the common feature space. Note that the social graph creation unit 103 predicts and creates implicit links between users by performing learning (representation learning, relationship learning, embedding learning, knowledge graph embedding) of a user relationship graph composed of nodes (users) connected by explicit links. At this time, the social graph creation unit 103 may perform such learning based on a known embedding model or its extension as appropriate.

[0051] <S42: Inference of Relationship between Links> In S42, the social graph creation unit 103 infers the relationship between the links predicted and created in S41. The inference process of the relationship between links will be described with reference to FIGS. 6A and 6B. FIG. 6A is a diagram for explaining the inference process of the relationship between links, and shows an example of inferring the relationship of a link between user A and user B connected by an explicit link.

[0052] The social graph creation unit 103 treats pairs of users connected by the links created in S41 as data points, and groups the pairs (data points) into clusters that represent a common type using various information acquired by the user feature acquisition unit 101. The various information may be information such as IP address, address, credit card, age, gender, and friends. Each cluster may have a relationship such as spouse, parent and child, neighbor, same household, colleague, friend, same-gender sibling, and different-gender sibling. In the example of FIG. 6A, user pairs are indicated by crosses, and a parent-child cluster 61, a spouse cluster 62, a same-gender sibling cluster 63, a friend cluster 64, and a same-household cluster 65 are shown as clusters into which the pairs may be grouped. Note that, although five clusters are shown in FIG. 6A, the number of clusters is not limited to a specific number.

[0053] For example, if user A and user B have (share) the following characteristics 60: the same surname, an age difference of less than 10 years, opposite gender, and the same address, the social graph creation unit 103 can group the pair of user A and user B into a cluster (spouse cluster 62) representing the relationship of husband and wife (spouse).

[0054] FIG. 6B shows a flowchart of an example of a process executed by the social graph creation unit 103 for grouping pairs into clusters. At the start of S61, for pairs to be grouped, assume they have the characteristics of the same address and the same surname. In S62, the social graph creation unit 103 determines whether the pair in question has the characteristic of the same gender. If the pair in question has the characteristic of the same gender (Yes in S62), in S63, the social graph creation unit 103 determines whether the age difference of the pair in question is less than or equal to a predetermined threshold value (= X value). If the age difference of the pair in question is greater than the X value (No in S63), the social graph creation unit 103 groups the pair in question into the parent - child cluster 61. If the age difference is less than or equal to the X value (Yes in S63), the social graph creation unit 103 groups the pair in question into the same - gender sibling cluster 63. Also, if the pair in question does not have the characteristic of the same gender (No in S62), in S64, the social graph creation unit 103 determines whether the age difference of the pair in question is less than or equal to a predetermined threshold value (= Y value). If the age difference is greater than the Y value (No in S64), the social graph creation unit 103 groups the pair in question into the parent - child cluster 61. If the age difference is less than or equal to the Y value (Yes in S64), the social graph creation unit 103 groups the pair in question into the spouse cluster 62.

[0055] <S43: Assignment of Affinity Score Based on Relational Affinity> In S43, the social graph creation unit 103 predicts a score (a score representing the degree of social relatedness) based on the relational affinity for the pairs inferred in S42, and assigns the score to the pairs. In this embodiment, this score is referred to as the affinity score. In this embodiment, the affinity score is a numerical value between 0 and 1, but there is no specific limitation on the numerical values that the affinity score can take. FIG. 7A shows a conceptual diagram of the affinity score based on the relational affinity for user pairs.

[0056] In the example of FIG. 7A, the intimacy of the relationship between the user pair changes depending on the characteristics shared by user A and user B who are connected by an explicit link. In the upper part of FIG. 7A, if user A and user B have characteristics 70, such as siblings of the same gender, the same address, 1200 call histories, and 50 gift exchanges, the intimacy of the relationship between the user pair (i.e., the intimacy score) is high. On the other hand, in the lower part of FIG. 7A, if user A and user B have characteristics 71, such as siblings of the same gender, different addresses, 30 call histories, and 2 gift exchanges, the intimacy of the relationship between the user pair (i.e., the intimacy score) is low. Thus, as in the example of FIG. 7A, even if user A and user B are siblings of the same gender, the intimacy of the relationship between the pair will differ depending on other characteristics shared by the user pair. It is observed that pairs with high intimacy of the relationship are close to each other in social distance and have high influence. On the other hand, pairs with low relationship intimacy are observed to have a large social distance from each other and are not close to each other.

[0057] In this embodiment, the intimacy score for a user pair is predicted using an intimacy score prediction model 112. Fig. 7B shows a schematic architecture of the intimacy score prediction model 112. The intimacy score prediction model 112 is a learning model that predicts an intimacy score 73 for a feature 72 of a user pair (a feature shared by the pair) as an input.

[0058] The intimacy score prediction model 112 is, for example, a learning model that performs weak supervised learning. In this embodiment, the intimacy score prediction model 112 is a learning model that is trained using intimacy scores (0 to 1) assigned to a plurality of features for a user pair as teacher data, as shown in FIG. 6A. For example, in the learning stage, combined data such as an intimacy score close to 1 set for feature 70 in FIG. 7A and an intimacy score close to 0 set for feature 71 are used as teacher data. Note that the intimacy score prediction model 112 may differ for each type of relationship between user pairs, and may be a learning model trained according to one type of relationship.

[0059] In this embodiment, the intimacy score for a user pair is predicted using the intimacy score prediction model 112, but the social graph creation unit 103 may be configured to predict the score using other methods. In this manner, the social graph creation unit 103 assigns an intimacy score to the user pair based on the characteristics shared by the pair.

[0060] Through the above process, explicit or implicit links are formed between multiple users, an intimacy score is assigned between each link, and a social graph is created. A conceptual diagram of a social graph is shown in Figure 7C. In the social graph 700 shown in Figure 7C, each of users 701 to 703 has multiple characteristics, and a predicted intimacy score is assigned to each pair of users as described above.

[0061] [Family user prediction procedure] Next, a procedure for predicting family users according to this embodiment will be described. In this embodiment, the family user prediction unit 104 predicts family users including registered users based on the social graph created by the social graph creation unit 103. In this embodiment, the family users are predicted based not only on the intimacy scores (see FIG. 7C) but also on the shared (common) user features between each pair in the social graph (see FIG. 6A).

[0062] The family user prediction unit 104 first determines whether or not there is a cohabiting user who lives with the registered user. For example, the family user prediction unit 104 searches for a user in the social graph who satisfies a first condition that the intimacy score in relation to the registered user is high (for example, the intimacy score is equal to or higher than a predetermined threshold) and has the same surname as the registered user and a common address. If a user who satisfies the first condition exists, it can be determined that a cohabiting user exists, and if not, it can be determined that a cohabiting user does not exist. The family user prediction unit 104 may also search for a user in the social graph who satisfies a second condition that the user has a common address with the registered user. In this case, if a user who satisfies the second condition exists, it can be determined that a cohabiting user exists, and if not, it can be determined that a cohabiting user does not exist.

[0063] If it is determined that there are no cohabiting users, the family user prediction unit 104 predicts that only registered users are family users. On the other hand, if it is determined that there are cohabiting users, it predicts that users who satisfy the conditions used for determining whether or not there are cohabiting users (for example, the above-mentioned first condition or second condition) are cohabiting users. Next, the family user prediction unit 104 extracts the relationship (family structure) between the registered user and the cohabiting user based on the social graph. In the example of FIG. 2, the family user prediction unit 104 recognizes that the spouse user 21 is the spouse of the registered user 20, and the child user 22 is the child of the registered user 20.

[0064] [Learning procedure for the viewing score prediction model] Next, a learning procedure of the viewing score prediction model 113 will be described with reference to Fig. 2 and Figs. 8A to 8C. Figs. 8A and 8B are diagrams for explaining the learning procedure of the viewing score prediction model 113. The viewing score prediction model 113 according to this embodiment is a tree (decision tree)-based model using boosting, for example, a learning model based on XGBoost. Note that the viewing score prediction model 113 may be learned individually or integrated for each time slot.

[0065] With reference to FIG. 8A, the relationship between the input data and the output data of the viewing score prediction model 113 will be described. In the example of FIG. 8A, the combined data of the content 26-1 and the user characteristics 80 is input to the viewing score prediction model 113. Here, the input content includes the content ID and metadata of the content. The user characteristics 80 also include demographic information, purchase history, and service usage status on the web service. The user characteristics 80 may be configured to include other types of user characteristics. For the input combined data, the viewing score 81 output from the viewing score prediction model 113 indicates the probability (possibility) that the input content 26-1 will be selected (including viewed) by a user having the input user characteristics 80.

[0066] During learning, the viewing score prediction model 113 is trained using combination data generated by the viewing user prediction unit 105. Since the combination data indicates a combination of viewed content and a viewing user, the viewing score output from the viewing score prediction model 113 is trained as 1 (maximum probability). For example, in FIG. 8A, combination data of content 26-1 and user feature 80 is input, and learning is performed to output a viewing score 81 (=1).

[0067] With reference to FIG. 8B, a learning process using combination data generated by the viewing user prediction unit 105 for the family user 23 shown in FIG. 2 will be described. With reference to FIG. 2, it is assumed that the family users 23 including the registered user 20 have viewed the content 26. In this case, the viewing user prediction unit 105 generates combination data of the user characteristics of at least one of the registered user 20, the spouse user 21, and the child user 22 and each of the contents 26-1 to 26-m. In the example of FIG. 8B, the content 26 includes the contents 26-1 to 26-9 (i.e., m=9). The generated combination data includes combination data of the contents 26-1 to 26-3 and the user characteristics 80 of the registered user 20, combination data of the contents 26-4 to 26-6 and the user characteristics 83 of the spouse user 21, and combination data of the contents 26-6 to 26-9 and the user characteristics 84 of the child user 22. The content 26-6 is combined with the user characteristics of the spouse user 21 and the child user 22. This means that the viewing users of the content 26-6 are predicted to be the spouse user 21 and the child user 22 by the viewing user prediction unit 105, and the content 26-6 is associated with both users.

[0068] Using the combination data, a viewing score prediction model 113 (user-based viewing score prediction model) is trained for each of the family users 23. In FIG. 8B, the viewing score prediction models 113 for the registered user 20, the spouse user 21, and the child user 22 are shown as viewing score prediction models 113-1, 113-2, and 113-3, respectively. The viewing score prediction model 113-1 is trained to receive combination data of each of the contents 26-1 to 26-3 and the user characteristic 82 of the registered user 20, and to output a viewing score 81-1 (=1). The viewing score prediction model 113-2 is trained to receive combination data of each of the contents 26-4 to 26-6 and the user characteristic 83 of the spouse user 21, and to output a viewing score 81-2 (=1). The viewing score prediction model 113-3 is trained to receive combination data of each of the contents 26-6 to 26-9 and the user characteristics 84 of the child user 22 and to output the viewing score 81-3 (=1).

[0069] 8A, the user characteristics 82 to 84 are configured to include, but are not limited to, demographic information, purchase history, and service usage status on the web service. In addition, the user characteristics 83 of the spouse user 21 and the user characteristics 84 of the child user 22 may include the relationship in the family structure with the registered user 20. For example, the user characteristics 83 may include a characteristic indicating "spouse," and the user characteristics 84 may include a characteristic indicating "child."

[0070] Furthermore, when content 26-6 to content 26-9 include information on the viewing date and time of the content, viewing score prediction models 113-1 to 113-3 may be trained to perform weighting according to the viewing date and time. For example, viewing score prediction models 113-1 to 113-3 may be trained so that the closer the viewing date and time is to the present or a predetermined time period, the higher the viewing score. This may increase the possibility that content having metadata similar to that of a recently viewed content will be determined as recommended content.

[0071] The learned viewing score prediction models 113-1 to 113-3 are integrated into a viewing score prediction model 113 for all family users. FIG. 8B shows a conceptual diagram of the viewing score prediction model 113 obtained by integrating the learned viewing score prediction models 113-1 to 113-3. By using the integrated viewing score prediction model 113, it is possible to predict the viewing scores of all family users 23 for the input content (including metadata and content ID). The learning unit 106 stores the viewing score prediction model 113 generated by the integration in the learning model storage unit 110 as the learned viewing score prediction model 113. The learning unit 106 may also store the viewing score prediction models 113 learned for each user (for example, the viewing score prediction models 113-1 to 113-3 in FIG. 8B) in the learning model storage unit 110.

[0072] When the family user prediction unit 104 determines that there are cohabiting users, for example, as shown in Fig. 8B, the viewing score prediction models 113-1 to 113-3 are trained using combination data of each family user and each content. On the other hand, when it is predicted that there are no cohabiting users, all the viewed content is considered to have been viewed by registered users, and more training data can be used for training. Therefore, the family user prediction unit 104 may preferentially train the viewing score prediction models 113 for registered users predicted to have no cohabiting users.

[0073] For example, with reference to Fig. 2, assume that registered user 20 has viewed all of content 26. In this case, viewing user prediction unit 105 generates combination data of the user characteristics of registered user 20 and each of contents 26-1 to 26-m. Then, viewing score prediction model 113 receives the combination data of each of contents 26-1 to 26-m and the user characteristics of registered user 20 and is trained to output a viewing score (=1). By such a learning process for registered users only, the accuracy of viewing scores for users having the user characteristics of registered users can be improved.

[0074] [Recommendation content prediction procedure] Next, a procedure for determining recommended content by the recommended content determination unit 107 will be described with reference to Fig. 2, Fig. 9A, and Fig. 9B. Fig. 9A and Fig. 9B are diagrams for explaining the procedure for determining recommended content. In this embodiment, the recommended content determination unit 107 is configured to determine recommended content using the viewing score prediction model 113 that is learned and generated by the learning unit 106.

[0075] In Fig. 9A, contents 91 including contents 91-1 to 91-n (n>0) are recommendation candidate contents (candidate contents) prepared in advance. Contents 91 may be input by an operator via an input unit (input unit 1105 in Fig. 11) or via a communication unit (communication I / F 1107 in Fig. 11). Alternatively, contents 91 may be set in advance in the system or may be set by an arbitrary program stored in a storage unit (ROM 1102 or RAM 1103 in Fig. 11).

[0076] The recommended content determination unit 107 inputs the content 91 (including metadata and content ID) and the user characteristics of each family user 23 to the trained viewing score prediction model 113. Specifically, the recommended content determination unit 107 inputs combination data of each of the content 91 and the user characteristics 82 of the registered user 20, combination data of each of the content 91 and the user characteristics 83 of the spouse user 21, and combination data of each of the content 91 and the user characteristics 83 of the child user 22 to the viewing score prediction model 113. The viewing score prediction model 113 predicts and outputs the viewing score 92 of the entire family users 23 from the input data. The recommended content determination unit 107 can determine, among the contents 91-1 to 91-n, a content having a viewing score 92 equal to or greater than a predetermined threshold as a recommended content for the family user 23. The predetermined threshold is, for example, 0.7. A content having a higher viewing score 92 is more likely to be viewed by at least one of the family users 23, and the recommendation effect is high. Note that the content 91 may include at least a part of the content 26 used in the learning stage.

[0077] In this way, by using the viewing score prediction model 113, it is possible to determine recommended content for all family users. In addition to this, or instead of this, the recommended content determination unit 107 may be configured to determine recommended content for each user constituting the family users. With reference to Fig. 9B, a procedure for determining recommended content for each of the family users 23 in Fig. 2 will be described.

[0078] 9B, the recommended content determination unit 107 inputs the content 91 (including metadata and content ID) and the user characteristics of each of the family users 23 to the trained viewing score prediction models 113-1 to 113-3. Specifically, the recommended content determination unit 107 inputs combination data of each of the content 91 and the user characteristics 82 of the registered user 20 to the viewing score prediction model 113-1. Similarly, the recommended content determination unit 107 inputs combination data of each of the content 91 and the user characteristics 83 of the spouse user 21 to the viewing score prediction model 113-2, and inputs combination data of each of the content 91 and the user characteristics 83 of the child user 22 to the viewing score prediction model 113-3. The viewing score prediction models 113-1 to 113-3 predict and output the viewing score 92-1 of the registered user 20, the viewing score 92-2 of the spouse user 21, and the viewing score 92-3 of the child user 22 from the input data. The recommended content determination unit 107 can determine, among the contents 91-1 to 91-n, the content having a viewing score equal to or greater than a predetermined threshold (for example, 0.7) as the recommended content. In the example of FIG. 8B, the recommended content determination unit 107 can determine, as the recommended content for the registered user 20, the content having the viewing score 92-1 equal to or greater than a predetermined threshold. Similarly, the recommended content determination unit 107 can determine, as the recommended content for the spouse user 21, the content having the viewing score 92-2 equal to or greater than a predetermined threshold ... The content 91 may include at least a part of the content 26 used in the learning stage.

[0079] Furthermore, the recommended content determination unit 107 may determine a viewing score 93 for all the family users 23 from the viewing scores 92-1 to 92-3. For example, the recommended content determination unit 107 may calculate the sum of the viewing scores 92-1 to 92-3 for each piece of content, and determine the sum as the viewing score 93. Alternatively, the recommended content determination unit 107 may extract the maximum score of each of the viewing scores 92-1 to 92-3 for each piece of content, and determine the extracted three maximum scores as the viewing score 93. Then, the recommended content determination unit 107 can determine a piece of content having a viewing score 93 equal to or greater than a predetermined threshold as recommended content for all the family users 23.

[0080] The recommended content for all family users 23 determined by the recommended content determination unit 107 is displayed on the communication device 11 (for example, smart TV 251 or tablet terminal 252 in FIG. 2) of the registered user 20 by the output unit 108. Here, the learning unit 106 can further train the viewing score prediction model 113 (and / or the viewing score prediction models 113-1 to 113-3) using a conversion rate (CVR) based on whether the content was viewed by any of the family users.

[0081] [Overall process flow] 10 shows a flowchart of the overall process executed by the information processing device 10 according to this embodiment. For the explanation of FIG. 10, the information processing system shown in FIG.

[0082] In S1001, user feature acquisition unit 101 acquires user features of each user from communication devices 11-1 to 11-N, and stores them in data storage unit 120 as user features 121. The process of S1001 may be an acquisition (collection) process of user features for a certain period of time in the past and estimated user features.

[0083] In S1002, the viewing history acquisition unit 102 acquires content viewing history information from the communication devices 11 of registered users who are registered with the content distribution service among the users 1 to N. The viewing history information includes information on the content ID, metadata, and viewing date and time (viewing time) of each content that has been viewed. The viewing history acquisition unit 102 stores the acquired viewing history information in the data storage unit 120 as the viewing history 122.

[0084] In S1003, the social graph creation unit 103 creates a social graph for users 1 to N using the user characteristics 121 stored in the data storage unit 120 in S1001. The social graph creation unit 103 assigns an intimacy score to each user pair in the social graph based on the characteristics common to the pair. The specific procedure for creating the social graph is as described above.

[0085] In S1004, the family user prediction unit 104 identifies registered users of the content distribution service, and predicts family users including the registered users. The family user prediction unit 104 acquires the user ID of the input (set) registered user, and acquires user features (user features of the registered user) corresponding to the user ID of the registered user from the user features 121. Next, the family user prediction unit 104 predicts cohabiting users who live with the registered user based on the acquired user features of the registered user and the social graph created in S1003, and predicts family users. The specific prediction procedure for family users is as described above.

[0086] In S1005, the viewing user prediction unit 105 predicts (identifies) viewing users of one or more contents included in the viewing history based on the viewing history of the registered user included in the viewing history 122 stored in the data storage unit 120 in S1002 and the family users predicted in S1004. The prediction corresponds to an association (alignment) between each of the one or more contents and each of the family users. Furthermore, the viewing user prediction unit 105 generates combination data between each of the contents included in the viewing history 122 and one or more viewing users from the association. The combination data includes the content ID and metadata of the content, and the user characteristics of the viewing data associated with the content. The combination data may also include information on the viewing date and time of the content.

[0087] In S1006, the learning unit 106 trains the viewing score prediction model 113. Specifically, the learning unit 106 first trains the viewing score prediction model 113 corresponding to each viewing user based on the combination data of the content and the viewing user generated in S1005. In the example of FIG. 8B, the learning unit 106 trains the viewing score prediction models 113-1 to 113-3. Furthermore, the learning unit 106 generates the viewing score prediction model 113 for all family users by integrating the viewing score prediction models 113 corresponding to each viewing user. In the example of FIG. 8C, the learning unit 106 generates the viewing score prediction model 113 by integrating the viewing score prediction models 113-1 to 113-3. The specific learning and generation procedure of the viewing score prediction model 113 is as described above.

[0088] In S1007, the recommended content determination unit 107 determines the recommended content for the family users. The recommended content determination unit 107 determines the recommended content for the family users using the viewing score prediction model 113 trained in S1006. Here, the recommended content determination unit 107 may determine the recommended content for each family user. The specific procedure for determining the recommended content is as described above.

[0089] In S1008, the output unit 108 presents the recommended content determined in S1007 to the family users including the registered user. For example, referring to Fig. 2, the output unit 108 displays the recommended content for the family user 23 on the communication device 11 (for example, the smart TV 251 or the tablet terminal 252 in Fig. 2) of the registered user 20. Here, the recommended content may be displayed so that the content can be understood, such as a part of an image of the content, or the title and performers of the content.

[0090] [Hardware configuration of information processing device 10] FIG. 11 is a block diagram showing an example of a hardware configuration of an information processing device 10 according to the present embodiment. The information processing apparatus 10 according to the present embodiment can be implemented on any single or multiple computers, mobile devices, or any other processing platform. 11, an example in which the information processing device 10 is implemented in a single computer is shown, but the information processing device 10 according to the present embodiment may be implemented in a computer system including multiple computers. The multiple computers may be connected to each other so as to be able to communicate with each other via a wired or wireless network.

[0091] 11, the information processing device 10 may include a CPU 1101, a ROM 1102, a RAM 1103, a HDD 1104, an input unit 1105, a display unit 1106, a communication I / F 1107, and a system bus 1108. The information processing device 10 may also include an external memory. A CPU (Central Processing Unit) 1101 generally controls the operations of the information processing device 10, and controls each of the components (1102 to 1107) via a system bus 1108, which is a data transmission path.

[0092] The ROM (Read Only Memory) 1102 is a non-volatile memory that stores a control program and the like necessary for the CPU 1101 to execute processing. Note that the program may be stored in a non-volatile memory such as a HDD (Hard Disk Drive) 1104 or an SSD (Solid State Drive) or an external memory such as a removable storage medium (not shown). The RAM (Random Access Memory) 1103 is a volatile memory, and functions as a main memory, a work area, etc. of the CPU 1101. That is, when executing a process, the CPU 1101 loads necessary programs, etc. from the ROM 1102 into the RAM 1103, and executes the programs, etc. to realize various functional operations. The learning model storage unit 110 and the data storage unit 120 shown in FIG. 2 can be configured by the RAM 1103.

[0093] The HDD 1104 stores, for example, various data and various information required when the CPU 1101 performs processing using a program. The HDD 1104 also stores, for example, various data and various information obtained when the CPU 1101 performs processing using a program. The input unit 1105 is composed of a keyboard and a pointing device such as a mouse. The display unit 1106 is configured with a monitor such as a liquid crystal display (LCD), etc. The display unit 1106 may function as a GUI (Graphical User Interface) by being configured in combination with the input unit 1105.

[0094] The communication I / F 1107 is an interface that controls communication between the information processing device 10 and an external device. The communication I / F 1107 provides an interface with a network and executes communication with an external device via the network. Various data, various parameters, etc. are transmitted and received between the external device and the communication I / F 1107. In this embodiment, the communication I / F 1107 may execute communication via a wired LAN (Local Area Network) or a dedicated line conforming to a communication standard such as Ethernet (registered trademark). However, the network available in this embodiment is not limited to this, and may be configured as a wireless network. This wireless network includes wireless PANs (Personal Area Networks) such as Bluetooth (registered trademark), ZigBee (registered trademark), and UWB (Ultra Wide Band). It also includes wireless LANs (Local Area Networks) such as Wi-Fi (Wireless Fidelity) (registered trademark), and wireless MANs (Metropolitan Area Networks) such as WiMAX (registered trademark). It also includes wireless WANs (Wide Area Networks) such as LTE / 3G, 4G, and 5G. Note that the network only needs to connect each device so that they can communicate with each other, and the communication standard, scale, and configuration are not limited to those described above.

[0095] At least some of the functions of each element of the information processing device 10 shown in Fig. 3 can be realized by the CPU 1101 executing a program. However, at least some of the functions of each element of the information processing device 10 shown in Fig. 2 may be operated as dedicated hardware. In this case, the dedicated hardware operates under the control of the CPU 1101.

[0096] [Hardware configuration of communication device 11] The hardware configuration of the communication device 11 shown in Fig. 1 may be the same as that of Fig. 11. That is, the communication device 11 may include a CPU 1101, a ROM 1102, a RAM 1103, a HDD 1104, an input unit 1105, a display unit 1106, a communication I / F 1107, and a system bus 1108. The communication device 11 can display various information provided by the information processing device 10 on the display unit 1106, and perform processing corresponding to an input operation received from a user via a GUI (configured by the input unit 1105 and the display unit 1106).

[0097] In this manner, in this embodiment, family users including registered users are predicted from a social graph created based on the user characteristics of an unspecified number of users. Then, viewing users of each of a plurality of contents provided to the registered users (i.e., viewed by the registered users) are predicted (identified). A user-based viewing score prediction model 113 (first learning model) for each family user is trained by the combined data obtained by the prediction (identification), and a viewing score prediction model 113 (second learning model) for the entire family users 23 is generated by integrating the trained plurality of first learning models. The generated viewing score prediction model 113 is used to determine recommended content for the entire family users. The determined recommended content is content with a higher viewing score (probability that the content will be selected), which is the output of the viewing score prediction model 113. By presenting the recommended content to the family users, it is expected that the effectiveness of the recommendation, such as an improvement in CVR, will be improved.

[0098] Although a specific embodiment has been described above, the embodiment is merely an example and is not intended to limit the scope of the present invention. The device and method described in this specification can be embodied in forms other than those described above. Furthermore, the above-described embodiments can be appropriately omitted, substituted, and modified without departing from the scope of the present invention. Such omitted, substituted, and modified forms are included in the scope of the claims and their equivalents, and belong to the technical scope of the present invention.

[0099] The disclosure of this embodiment includes the following configuration. [1] An information processing device comprising: a graph creation means for creating a relationship graph showing social relationships between a plurality of users based on the user characteristics of the plurality of users; a prediction means for predicting family users among the plurality of users, including registered users who are registered with a content distribution service and cohabiting users who live with the registered users, using the relationship graph; an identification means for identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; and a decision means for determining recommended content to be recommended to the family user using a learning model based on the results of the identification.

[0100] [2] The information processing device described in [1], wherein the learning model is configured to input metadata representing the characteristics of candidate content and user characteristics of the family user, and output the probability that the candidate content will be selected by the family user, and the determination means uses the learning model to determine the recommended content for the family user.

[0101] [3] The information processing device described in [1] or [2], wherein the user characteristics of each of the family users include behavioral tendencies of each of the family users, and the identification means identifies which of the family users has viewed each of the multiple contents in accordance with the behavioral tendencies of each of the family users.

[0102] [4] An information processing device described in any of [1] to [3], wherein the identification means generates, as a result of the identification, combination data for each of the plurality of contents, which combines metadata representing the characteristics of the content with the user characteristics of one or more users who are identified as having viewed the content.

[0103] [5] The information processing device described in [4] further comprises a learning means for generating the learning model, wherein the learning means uses the combination data to learn a first learning model for each of the family users, and generates a second learning model that integrates the first learning models of all the family users as the learning model.

[0104] [6] The information processing device according to any one of [1] to [5], wherein the identification means identifies which of the family users has viewed each of the plurality of contents on a rule-based basis.

[0105] [7] An information processing device described in any of [1] to [6], wherein the plurality of contents include at least one of cultural content including movies, anime, or dramas, and advertising content. [Explanation of symbols]

[0106] 1 to N: users, 10: information processing device, 11-1 to 11-N: communication devices, 101: user feature acquisition unit, 102: viewing history acquisition unit, 103: social graph creation unit, 104: family user prediction unit, 105: viewing user prediction unit, 106: learning unit, 107: recommended content determination unit, 108: output unit, 110: learning model storage unit, 111: user feature prediction model, 112: intimacy score prediction model, 112: viewing score prediction model, 120: data storage unit, 121: user features, 122: viewing history

Claims

1. A prediction means for predicting family users including registered users who are registered with a content distribution service and cohabiting users who live with the registered users, among the plurality of users, by using a relationship graph showing social relationships between the plurality of users; an identification means for identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; A determination means for determining a recommendation content to be recommended to the family user by a learning model based on the result of the identification; 13. An information processing device comprising:

2. the learning model is configured to receive metadata representing characteristics of the candidate content and the user characteristics of the family user, and output a probability that the candidate content will be selected by the family user; The determination means determines the recommended content for the family user using the learning model.

2. The information processing apparatus according to claim 1,

3. The user characteristics of each of the family users include a behavioral tendency of each of the family users; the identification means identifies which of the family users has viewed each of the plurality of contents in accordance with a behavioral tendency of each of the family users.

2. The information processing apparatus according to claim 1,

4. the identification means generates, as a result of the identification, combination data for each of the plurality of contents, the combination data being a combination of metadata representing a feature of the content and user features of one or more users identified as having viewed the content; 2. The information processing apparatus according to claim 1,

5. Further comprising a learning means for generating the learning model, The learning means uses the combination data to learn a first learning model for each of the family users; generating a second learning model by integrating the first learning models of all the family users as the learning model; 5. The information processing apparatus according to claim 4.

6. the identification means identifies, on a rule basis, which of the family users has viewed each of the plurality of contents; The information processing device according to claim 1 .

7. The plurality of contents include at least one of culture contents including movies, anime, or dramas, and advertising contents; 2. The information processing apparatus according to claim 1,

8. a prediction step of predicting, by using a relationship graph showing social relationships between a plurality of users, family users including registered users who are registered with a content distribution service and cohabiting users who live with the registered users, among the plurality of users; an identification step of identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; a determining step of determining, based on the result of the identification, a recommendation content to be recommended to the family user by a learning model; 23. An information processing method comprising:

9. An information processing program for causing a computer to execute information processing, the program comprising: A prediction process for predicting family users including registered users who are registered with a content distribution service and cohabiting users who live with the registered users, among the plurality of users, using a relationship graph showing social relationships between the plurality of users; an identification process for identifying which of the family users has viewed each of a plurality of contents provided to the registered user by the content distribution service; and a determination process of determining, based on a result of the identification, a recommendation content to be recommended to the family user using a learning model. Information processing program.

Images