Information processing device, information processing method, and recording medium

Abstract

One objective of the present disclosure is to provide an information processing device and the like that support the improvement of convenience in services that use information about user preferences. An information processing device according to one aspect of the present disclosure comprises: an extraction means for extracting target information, which is information about a target of an action of a predetermined user in an online service where information is posted; an identification means for identifying feature information indicating a feature related to the extracted target information; a generation means for generating attribute information indicating an attribute preferred by the predetermined user, on the basis of the identified feature information; and an output means for outputting basis information that is information relating to the predetermined user's action and that indicates the basis on which the attribute information for the predetermined user was generated.

Description

Information Processing Apparatus, Information Processing Method, and Recording Medium

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and a recording medium.

[0002] There is a technique for inferring a predetermined property regarding a user who uses a system.

[0003] For example, Patent Document 1 discloses a technique for estimating the preferences of a newly registered user based on posting information of others that the newly registered user empathized with in a web application.

[0004] International Publication No. 2022 / 185401

[0005] Information indicating a predetermined property regarding a user may be used for services for the user. For example, it is conceivable to recommend products according to the user's preferences using information indicating the user's preferences. In Patent Document 1, information indicating the estimated preferences of a newly registered user is used for matching between users.

[0006] Thus, information on user preferences can be considered to be used in various services. Improvement in convenience in such services is required.

[0007] One of the objects of the present disclosure is to provide an information processing apparatus or the like that supports improvement in convenience in services using information regarding user preferences.

[0008] An information processing apparatus according to an aspect of the present disclosure includes: extraction means for extracting target information that is information regarding a target of an action of a predetermined user in an online service where information is posted; specification means for specifying feature information indicating features related to the extracted target information; generation means for generating attribute information indicating an attribute that the predetermined user prefers based on the specified feature information; and output means for outputting basis information that is information regarding the action of the predetermined user and indicates a basis for generating the attribute information for the predetermined user.

[0009] An information processing method according to one aspect of this disclosure extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted; identifies feature information that shows characteristics related to the extracted target information; generates attribute information that shows the attributes preferred by the predetermined user based on the identified feature information; and outputs basis information that shows the basis for the generation of attribute information for the predetermined user, relating to the actions of the predetermined user.

[0010] A recording medium according to one aspect of this disclosure stores a program that causes a computer to execute the following: a process for extracting target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted; a process for identifying feature information that indicates features related to the extracted target information; a process for generating attribute information that indicates attributes preferred by the predetermined user based on the identified feature information; and a process for outputting basis information that indicates the basis for the generation of attribute information for the predetermined user, which is information relating to the actions of the predetermined user.

[0011] This disclosure can help improve the convenience of services that utilize information about user preferences.

[0012] This is a schematic diagram showing an example of a configuration including the information processing device of the present disclosure. This is a first block diagram showing an example of the functional configuration of the information processing device of the present disclosure. This is a first flowchart explaining an example of the operation of the information processing device of the present disclosure. This is a second block diagram showing an example of the functional configuration of the information processing device of the present disclosure. This is a diagram showing an example of registration information of the present disclosure. This is a diagram showing an example of output information of the present disclosure. This is a second flowchart explaining an example of the operation of the information processing device of the present disclosure. This is a third block diagram showing an example of the functional configuration of the information processing device of the present disclosure. This is a third flowchart explaining an example of the operation of the information processing device of the present disclosure. This is a block diagram showing an example of the hardware configuration of a computer device that implements the information processing device of the present disclosure.

[0013] Embodiments of this disclosure will be described below with reference to the drawings.

[0014] <First Embodiment> An overview of the information processing device of the first embodiment will be described.

[0015] Figure 1 is a schematic diagram showing an example of a configuration including an information processing device 100. The information processing system 1000 comprises the information processing device 100. In the example of Figure 1, the information processing system 1000 further comprises a service provision server 200. Note that the configuration of the information processing system 1000 is not limited to this example. For example, the information processing system 1000 may further comprise a user terminal 300.

[0016] The information processing device 100 may be connected to the service provision server 200 and the user terminal 300 via a wireless or wired network. The information processing device 100 may be able to communicate with other devices not shown. For example, the information processing device 100 may be connected to a device that outputs information, such as a terminal device. The information processing device 100 is, for example, a server device, but is not limited to this example. The information processing device 100 may be implemented as a cloud server. Furthermore, the information processing device 100 may be constructed from multiple devices.

[0017] The service provider server 200 is a device that provides an online service on which information is posted. The online service may be a service that allows users to interact with each other by forming a community on the internet and posting information. For example, the online service is an SNS (Social Networking Service). In other words, the service provider server 200 may be a server that provides an SNS platform. The service provider server 200 may be constructed by multiple devices.

[0018] Users of the online service access the service via a user terminal 300. The user terminal 300 is a terminal device such as a smartphone, tablet, or personal computer. Hereafter, users of the online service will also be simply referred to as "users."

[0019] An example of the information processing device 100 disclosed herein is a device that generates predetermined information for a user of an online service. The information processing device 100 is a device that outputs information related to the generated predetermined information. For example, the information processing device 100 may be capable of associating predetermined information with information about a user. The information processing device 100 may then display the associated predetermined information on a display device such as a display.

[0020] Next, an example of the functional configuration of the information processing device 100 will be described.

[0021] Figure 2 is a block diagram showing an example of the functional configuration of the information processing device 100. The information processing device 100 comprises an extraction unit 110, a specification unit 120, a generation unit 130, and an output unit 140.

[0022] The extraction unit 110 extracts target information from the online service. The target information is information related to the actions of a predetermined user in the online service. An example of target information is posted information. Posted information refers to information posted in the online service. Users may perform various reactions to posted information. Reactions include, for example, expressing empathy ("like", etc.), reposting, quoting, and bookmarking. The extraction unit 110 may, for example, extract posted information to which a predetermined user has reacted as target information.

[0023] Another example of target information is user information. User information refers to information about a user. Communities can be formed in online services when users follow other users or when users are followed by other users. The extraction unit 110 may use such user relationships to extract target information. For example, the extraction unit 110 may extract user information about other users who have a relationship with a given user as target information.

[0024] In this way, the extraction unit 110 extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted. The extraction unit 110 is an example of an extraction means.

[0025] The identification unit 120 identifies characteristic information related to the target information. Characteristic information is information that indicates features related to the target information. An example of characteristic information is attribute information pre-assigned to the target information. Attribute information includes information that indicates keywords related to the target information. Another example of characteristic information may be information that indicates features inferred from the target information. For example, suppose the posted information includes images and text. In this case, the characteristic information may be keywords included in the text and keywords related to objects detected from the image. The identification unit 120 identifies characteristic information related to the target information extracted by the extraction unit 110. Specifically, suppose the posted information includes images showing clothing products. In this case, the posted information may be associated with keywords indicating the type of clothing, the style of clothing, and the human sensibility that perceives clothing. For example, attribute information indicating keywords such as "dress," "formal," and "elegant" may be associated with the posted information. In this case, the identification unit 120 identifies "dress," "formal," and "elegant" as characteristic information related to the posted information. Note that the characteristic information is not limited to this example.

[0026] In this way, the identification unit 120 identifies feature information that indicates features related to the extracted target information. The identification unit 120 is an example of an identification means.

[0027] The generation unit 130 generates attribute information for a user. Specifically, the generation unit 130 generates attribute information for a predetermined user using the feature information identified by the identification unit 120. At this time, the generation unit 130 may generate attribute information that indicates the features shown in the identified feature information. For example, suppose the feature information includes the keyword "elegant". At this time, the generation unit 130 may generate attribute information for the predetermined user that indicates the keyword "elegant". Thus, the attribute information may include keywords that indicate sensibilities. Note that the method of generating attribute information is not limited to this example. For example, the generation unit 130 may generate attribute information that indicates the features shown in the feature information among the identified feature information where the number is above a threshold.

[0028] The target of an action performed by a designated user is highly likely to be something that the designated user is interested in. Characteristic information about such a target, which is information about that target, can be considered as attributes preferred by the designated user. In other words, the attribute information generated by the generation unit 130 indicates attributes preferred by the designated user.

[0029] In this way, the generation unit 130 generates attribute information indicating the attributes preferred by a predetermined user, based on the identified characteristic information. The generation unit 130 is an example of a generation means.

[0030] The output unit 140 outputs various types of information. For example, the output unit 140 outputs basis information indicating the reason for the generation of attribute information. The basis information includes information about the actions of a predetermined user. For example, the basis information includes information indicating the actions of a predetermined user with respect to target information to which the feature information used for generating the attribute information has been attached. The basis information is not limited to this example. The output unit 140 may also output information about the generated attributes.

[0031] The output unit 140 outputs various information to a predetermined device. Specifically, the output unit 140 may output supporting information to the user terminal 300. This allows the supporting information to be displayed on the user terminal 300, for example. The output unit 140 may also output supporting information to the service provision server 200. This allows the supporting information to be displayed on the user terminal 300 via the service provision server 200, for example. The output unit 140 is not limited to these examples and may output supporting information to other terminal devices not shown.

[0032] Thus, the output unit 140 outputs information relating to the actions of a predetermined user, and provides supporting information indicating the basis for the generation of attribute information for that predetermined user. The output unit 140 is an example of an output means.

[0033] Next, an example of the operation of the information processing device 100 will be explained using Figure 3. Figure 3 is a first flowchart illustrating an example of the operation of the information processing device 100. In this disclosure, each step of the flowchart is represented by a number assigned to each step, such as "S1".

[0034] The extraction unit 110 extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted (S1).

[0035] The identification unit 120 identifies feature information that indicates features related to the extracted target information (S2).

[0036] The generation unit 130 generates attribute information indicating the attributes preferred by a predetermined user based on the identified characteristic information (S3).

[0037] The output unit 140 outputs information relating to the actions of a predetermined user, which indicates the basis for the generation of attribute information for the predetermined user (S4).

[0038] Thus, the information processing device 100 of the first embodiment extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted. The information processing device 100 also identifies feature information that indicates characteristics related to the extracted target information. Furthermore, based on the identified feature information, the information processing device 100 generates attribute information indicating the attributes preferred by the predetermined user. Finally, the information processing device 100 outputs basis information, which is information relating to the predetermined user's actions and indicates the basis for the generation of attribute information for the predetermined user.

[0039] Various services may be provided using information that indicates the attributes a user prefers. For example, in online services such as social networking services (SNS), posts that correspond to the user's preferred attributes may be presented to that user. Also, for example, products that correspond to the user's preferred attributes may be recommended to that user. In such situations, the information processing device 100 can present the basis for generating the attribute information for the user.

[0040] This allows users, for example, to understand the basis for their attribute information. Knowing the basis for their attribute information allows users to understand, for example, what actions they should take to receive the desired tags.

[0041] In addition, a provider of a service using attribute information can manage, for example, the basis for attaching attribute information to a user. The provider of the service can also examine the validity of the attribute information attached to the user.

[0042] That is, the information processing apparatus 100 can support improvement of convenience in a service using information on a user's preferences.

[0043] <Second Embodiment> Next, an information processing apparatus according to the second embodiment will be described. In the second embodiment, a further example regarding the information processing apparatus 100 described in the first embodiment will be described. Note that descriptions of content overlapping with the first embodiment will be partially omitted.

[0044] The information processing apparatus 100 can communicate with a service providing server 200 that provides an online service and a user terminal 300 via a network.

[0045] An example of the information processing apparatus 100 is a part of a system that provides a service for providing appropriate information to a user. Specifically, the information processing apparatus 100 may be a server that realizes a portal site for recommending products to a user using SNS information. For example, a user registers with a portal site (that is, the information processing apparatus 100). The information processing apparatus 100 acquires various types of information in the SNS from the service providing server 200 and analyzes the actions of the registered user in the SNS. Further, the information processing apparatus 100 associates attribute information corresponding to the analysis result with the registered user. Here, the attribute information is also referred to as a "tag". Also, the process of associating such attribute information with a user is also referred to as "tagging". Then, the information processing apparatus 100 recommends products to the user according to, for example, the tags. In the present embodiment, an example in which the information processing apparatus 100 performs tagging on a user who uses the SNS will be described. Note that the application example of the information processing apparatus 100 is not limited to this example.

[0046] FIG. 4 is a block diagram showing an example of the functional configuration of the information processing apparatus 100. The information processing apparatus 100 includes an extraction unit 110, a specification unit 120, a generation unit 130, and an output unit 140. Further, the information processing apparatus 100 may include a determination unit 150 and an association unit 160. Furthermore, the information processing apparatus 100 may include a storage device 190. The storage device 190 may be a device possessed by the information processing apparatus 100, or may be an external device communicably connected to the information processing apparatus 100.

[0047] The information processing apparatus 100 associates attribute information with a target user. The target user is also referred to as a predetermined user. The predetermined user is, for example, a user registered on the above-described portal site. The information processing apparatus 100 has registration information, which is information about a user registered on the portal site. The registration information is, for example, a database including information about a user registered on the portal site. The registration information includes user identification information and SNS account information of the user. Note that the information included in the registration information is not limited to this example. For example, the registration information may include the user's date of birth, occupation, contact information, and the date and time when the user was registered. The registration information is stored in the storage device 190.

[0048] The determination unit 150 determines a user for whom attribute information is to be generated. For example, the determination unit 150 may determine the user as a predetermined user契机に when the user registers on the portal site. Further, the determination unit 150 may determine a user who has performed a predetermined action as a predetermined user. For example, the determination unit 150 may determine a user whose number of follows or followers on the SNS has reached a certain value as a predetermined user. Also, the determination unit 150 may determine a user whose number of posts on the SNS has reached a certain value as a predetermined user, or may determine a user whose elapsed days after registration on the SNS have reached a certain value as a predetermined user. The method of determining a predetermined user is not limited to this example.

[0049] Thus, the determination unit 150 may determine a user who satisfies a predetermined condition as a predetermined user.

[0050] The extraction unit 110 extracts target information for a predetermined user. Specifically, the extraction unit 110 extracts post information that has received a reaction from the predetermined user. A reaction indicates a user's response to a post. For example, when a user expresses empathy for a post on social media, they may press buttons such as "Like," "High Rating," "Favorite," and "Applause." Such expressions of empathy are examples of reactions. Reactions also include reposting and quoting posts. Furthermore, reactions include the number of times a post is viewed, bookmarking posts, and sharing posts with other applications. Note that reactions are not limited to these examples.

[0051] SNS posting information is managed, for example, on the service provider server 200. The extraction unit 110 may extract posting information that has received a reaction from a predetermined user from the service provider server 200. The extraction unit 110 may also extract all posting information that has received a reaction, or it may extract only a portion of the posting information that has received a reaction. For example, the extraction unit 110 may extract posting information that has received a reaction from a predetermined user within a certain period of time.

[0052] Thus, the target information may include posted information that has been posted on an online service. The extraction unit 110 may then extract posted information that a predetermined user has reacted to.

[0053] Furthermore, the extraction unit 110 extracts user information relating to other users who have a relationship with a predetermined user as target information. User information is managed, for example, on the service provision server 200. User information is information relating to SNS users. For example, user information may be information relating to a user's account. User information may include account name, account ID, profile, number of following, number of followers, and number of posts. The extraction unit 110 may extract user information from the service provision server 200. For example, the extraction unit 110 may extract user information of users that the predetermined user follows (i.e., followers). Alternatively, the extraction unit 110 may extract user information of users who follow the predetermined user (i.e., followers). Or, the extraction unit 110 may extract user information of users who are followers of the predetermined user and are also followers of the predetermined user. The method of extracting target information is not limited to the examples described above.

[0054] Thus, the target information may include user information, which is information about users in an online service. The extraction unit 110 may then extract user information about other users who have a relationship with a predetermined user.

[0055] The identification unit 120 identifies characteristic information related to the target information. An example of characteristic information is attribute information attached to the target information. For example, attribute information is associated with each post. That is, the post information is tagged in advance. The identification unit 120 may identify the attribute information associated with the extracted post information as characteristic information. For example, suppose the extracted post information is associated with attribute information indicating keywords such as "one-piece dress," "elegant," and "20s." In this case, the identification unit 120 identifies the keywords "one-piece dress," "elegant," and "20s."

[0056] Similarly, assume that attribute information is associated with user information. That is, user information is pre-tagged. The identification unit 120 may identify the information associated with the extracted user information as characteristic information.

[0057] The identification unit 120 may also identify information that indicates features inferred from the target information as feature information, although this is not limited to this example. The posted information includes at least one of an image and text. The identification unit 120 may identify features that are estimated from at least one of the image and text included in the posted information as feature information. Specifically, the identification unit 120 detects objects and people in the image and estimates their features. For example, suppose the image shows a woman in her 20s wearing a dress. The identification unit 120 uses existing image processing technology to detect the person and estimate the gender, age, and clothing of the detected person. The identification unit 120 may then identify keywords indicating the estimated gender, age, and clothing as feature information. The identification unit 120 may also identify keywords included in the text included in the posted information. For example, the identification unit 120 may use words that are used in the text more than a certain number of times as keywords. In this case, the identification unit 120 may use existing language processing technology to estimate the topic of the text and identify keywords indicating the estimated topic as feature information. Furthermore, if the text contains labels used on social media, such as hashtags, the identification unit 120 may identify such labels as feature information. The identification unit 120 may identify multiple such feature information.

[0058] In this way, the identification unit 120 may identify feature information that indicates features related to the extracted post information. At this time, the identification unit 120 may identify at least one of the following as feature information: attribute information attached to the extracted post information and keywords related to the content of the post information that are estimated from at least one of the images and text included in the post information.

[0059] Furthermore, the identification unit 120 may identify attribute information attached to the extracted user information as feature information. The identification unit 120 may also identify information estimated from the user information as feature information. For example, user information includes a profile. The identification unit 120 may identify keywords and labels used in the profile as feature information.

[0060] The generation unit 130 generates attribute information for a predetermined user using the identified feature information. The method for generating attribute information can be carried out in various ways. Specifically, the generation unit 130 may generate attribute information for a predetermined user by selecting the feature information that appears most frequently from among the identified feature information. For example, suppose the identified feature information includes the attributes "pretty," "cute," "elegant," "dress," "cap," and "short hair." The generation unit 130 may identify the top four attributes that appear most frequently from among these. For example, suppose "cute," "elegant," "dress," and "short hair" are identified. In this case, the generation unit 130 may generate attribute information indicating "cute," "elegant," "dress," and "short hair" as attribute information for a predetermined user.

[0061] Thus, the generation unit 130 may generate attribute information for a predetermined user from among the multiple identified feature information, where the number of feature information of the same type is above a threshold. It can be inferred that feature information with a large number of identified features indicates information that is of more interest to the predetermined user. In other words, the information processing device 100 can generate attribute information based on features that are of more interest to the predetermined user.

[0062] Furthermore, the generation unit 130 may generate attribute information for a predetermined user using a learning model. The learning model is a learning model that has learned the relationship between feature information identified from target information on which an action has been taken by the user, and attribute information indicating the attributes preferred by the user. For example, the learning model may be a model that outputs which of the identified feature information is most likely to be an attribute preferred by the user. That is, the learning model may be a model that takes identified feature information as input and outputs attribute information. The learning model may be generated by various machine learning algorithms such as regression analysis, support vector machines, neural networks, decision trees, and random forests. The learning model may be generated by the generation unit 130, or by other functional units not shown. Alternatively, the learning model may be generated by other devices that can communicate with the information processing device 100.

[0063] The learning model may be one that takes into account the weight of feature information. For example, the learning model may be one that tends to output feature information with a larger weight as an attribute that is more likely to be preferred by the user. Specifically, feature information that is identified in larger numbers may be given a larger weight. Also, if the post information that received a reaction is a quoted post, the weight of the feature information identified from the quoted post may be set lower or higher than the weight of the post information that is not a quoted post. Also, if the post information that received a reaction is an advertisement, the weight of the feature information identified from the advertisement may be set lower or higher than the weight of the post information that is not an advertisement. Furthermore, the weight of the feature information identified from the post information may be set lower or higher depending on whether the poster of the post information that received a reaction is an individual or a company. In this way, the feature information input to the learning model may be weighted according to at least one of the type of post information and the poster.

[0064] If the feature information is identified from the user information of other users who have a relationship with a given user, different weights may be set depending on whether the other user is a follower or a follower of the given user. For example, if the feature information is identified from the user information of a follower, a higher weight may be set compared to the feature information identified from the user information of a follower. Also, weights may be set according to the number of times the given user has reacted to other users' posts. For example, suppose there are multiple other users who have a relationship with the given user. In this case, the weight of the feature information identified from the user information of a user who has expressed empathy for a large number of posts may be set higher than the weight of the feature information identified from the user information of a user who has expressed empathy for a small number of posts. In this way, the feature information input to the learning model may be weighted according to the number of reactions to other users' posts. Note that the method of setting weights and the learning model are not limited to the examples above.

[0065] Thus, the identified feature information may be weighted according to the properties of the target information corresponding to the feature information. For example, the identified feature information may be weighted according to at least one of the type of posted information corresponding to the feature information and the poster of the posted information. Alternatively, for example, the identified feature information may be weighted according to the reaction of a predetermined user to posted information by other users corresponding to the feature information. The generation unit 130 may then generate attribute information for the predetermined user according to the weights of the identified feature information.

[0066] The association unit 160 associates the generated attribute information with a predetermined user. Specifically, the association unit 160 associates the generated attribute information with information about a predetermined user among the registered information. Figure 5 shows an example of registered information. In the example in Figure 5, the registered information is associated with a user ID, address, service ID, and attribute information. Here, the user ID is an example of user identification information on the portal site. The address is an example of user contact information. The service ID is an example of SNS account information used by the user. For example, the user in the first record has a user ID of "user-123" and an address of "aa@xxx.com". It is also shown that this user has an account with the ID "@abc123" on an SNS called "SNS-A" and an account with the ID "@abc_123" on an SNS called "SNS-B". The attribute information associated with this user is "elegant", "dress", "pink", and "short hair". Thus, the association unit 160 may associate users with attribute information by registering attribute information generated for each user in the registration information.

[0067] The association unit 160 may associate a user with attribute information by registering attribute information with the user information of a predetermined user. User information is, for example, information about a user's account on a social networking service (SNS). In other words, the association unit 160 may associate attribute information with information managed by the service provision server 200.

[0068] The output unit 140 outputs output information that includes the generated attribute information and the rationale information. That is, when a predetermined user is tagged, the output unit 140 may output information indicating the tag and information indicating why that tag was assigned.

[0069] The output unit 140 displays the output information on the user terminal 300, for example, by outputting the output information to the user terminal 300. Figure 6 is a diagram showing an example of the output information. More specifically, Figure 6 is a diagram showing an example of the output information to be displayed on the user terminal 300. For example, a user accesses the portal site via the user terminal 300. That is, the user accesses the information processing device 100. Then, the user's registration information is displayed based on the user's operation. In other words, the output unit 140 may display the output information by receiving user operations via the user terminal 300. In the example of Figure 6, information regarding the assigned tags (i.e., attribute information generated for the user) from the user's registration information is displayed. Specifically, it is shown that the user with user ID "user-123" has been assigned the tags "elegant", "one-piece dress", "pink", and "short hair". Furthermore, supporting information is shown for each tag. For example, for the tag "#elegant", the number of reactions to posts tagged with "#elegant" is shown as supporting information. Furthermore, for example, the tag "#shorthair" is based on the fact that the user follows users A, B, and C. In this case, it is possible that users A, B, and C have been tagged with "#shorthair". In other words, the basis information includes information indicating which action of the predetermined user is the basis for the characteristic information corresponding to the generated attribute information. In this way, the output unit 140 can output as basis information information information about the predetermined user's actions that were considered for generating attribute information for the predetermined user.

[0070] Furthermore, the output unit 140 may output information that accepts modification of attribute information. In the example in Figure 6, the output information includes a button labeled "Modify Tag". For example, when the user selects the "Modify Tag" button, the output unit 140 outputs output information that includes information that accepts input of attribute information. The user then modifies the tags. For example, the user deletes the tag "pink" from the generated tags. The output unit 140 accepts the tag modification from the user via the output information. At this time, the association unit 160 updates the information regarding the assigned tags in the user's registration information according to the accepted modification operation. For example, the association unit 160 deletes the tag "pink" that was assigned to the user. In this way, the output unit 140 may accept modification of attribute information. This allows the user to modify attribute information if there is attribute information that does not match their preferences.

[0071] Furthermore, the output unit 140 may output output information to other terminal devices that are communicatively connected to the information processing device 100. For example, the output unit 140 may output output information to a terminal device managed by a company operating a portal site. That is, the output unit 140 may output output information to a terminal device operated by the administrator of the portal site.

[0072] The destination of the output information is not limited to this example. For example, the output unit 140 may output the output information to a terminal device managed by a company that provides an SNS platform.

[0073] Thus, the output unit 140 may output attribute information for a predetermined user and supporting information for each attribute piece of information. In this case, the supporting information may include information indicating which action of the predetermined user the characteristic information corresponding to the generated attribute information is based on.

[0074] [Example of operation of the information processing device 100] Next, an example of the operation of the information processing device 100 will be explained using Figure 7.

[0075] Figure 7 is a second flowchart illustrating an example of the operation of the information processing device 100. First, the determination unit 150 determines a predetermined user (S101). For example, the determination unit 150 determines a user who meets predetermined conditions as a predetermined user. Specifically, the determination unit 150 may determine a predetermined user to be a user among the users registered on the portal site whose number of followers or followings on SNS has reached a certain value.

[0076] The extraction unit 110 extracts target information about a predetermined user (S102). For example, the extraction unit 110 extracts post information that has been reacted to by the predetermined user as target information. Alternatively, for example, the extraction unit 110 extracts user information about other users who have a relationship with the predetermined user as target information. The identification unit 120 identifies characteristic information related to the extracted target information (S103). For example, the identification unit 120 identifies attribute information attached to the target information.

[0077] The generation unit 130 generates attribute information for a predetermined user based on the identified feature information (S104). For example, the generation unit 130 may generate attribute information for a predetermined user by selecting feature information from the identified feature information whose identified number is greater than or equal to a threshold. Alternatively, for example, the generation unit 130 may generate attribute information for a predetermined user by utilizing a learning model that has learned the relationship between feature information identified from target information on which an action has been taken by the user and attribute information indicating the attributes preferred by that user.

[0078] The association unit 160 associates the generated attribute information with a predetermined user (S105). For example, the association unit 160 associates the generated attribute information with information about a predetermined user among the registered information.

[0079] The output unit 140 outputs output information (S106). For example, the output unit 140 displays the output information by outputting it to the user terminal 300 or to a terminal device managed by the company operating the portal site.

[0080] This example of operation is merely one example. In other words, the operation of the information processing device 100 in this disclosure is not limited to this example of operation.

[0081] Thus, the information processing device 100 of the second embodiment extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted. The information processing device 100 also identifies feature information that indicates characteristics related to the extracted target information. Furthermore, based on the identified feature information, the information processing device 100 generates attribute information indicating the attributes preferred by the predetermined user. Finally, the information processing device 100 outputs basis information, which is information relating to the predetermined user's actions and indicates the basis for the generation of attribute information for the predetermined user.

[0082] Various services may be provided using information that indicates the attributes a user prefers. For example, in online services such as social networking services (SNS), posts that correspond to the user's preferred attributes may be presented to that user. Also, for example, products that correspond to the user's preferred attributes may be recommended to that user. In such situations, the information processing device 100 can present the basis for generating the attribute information for the user.

[0083] This allows users, for example, to understand the basis for their attribute information. Knowing the basis for their attribute information allows users to understand, for example, what actions they should take to receive the desired tags. Furthermore, service providers using attribute information can manage, for example, what the basis is for assigning attribute information to users. Service providers can also review the legitimacy of the attribute information assigned to users.

[0084] In other words, the information processing device 100 can support the improvement of convenience in services that utilize information about user preferences.

[0085] The target information includes, for example, posted information that has been posted on an online service. The information processing device 100 may then extract posted information that a predetermined user has reacted to, and identify characteristic information that indicates features related to the extracted posted information.

[0086] Furthermore, the posted information includes, for example, at least one of an image and text. The information processing device 100 may then identify at least one of the following as feature information: attribute information attached to the extracted posted information and keywords related to the content of the posted information, which are estimated from at least one of the image and text included in the posted information.

[0087] Posts to which a designated user has reacted are presumed to contain content of interest to that designated user. In other words, the information processing device 100 can generate attribute information based on characteristic information corresponding to the post information of interest to the designated user.

[0088] Furthermore, the target information includes, for example, user information, which is information about users in an online service. The information processing device 100 may then extract user information about other users who have a relationship with a predetermined user, and identify the attribute information attached to the extracted user information as characteristic information.

[0089] Other users who have a relationship with a given user, such as followers and followings, are presumed to be users of interest to the given user. In other words, the information processing device 100 can generate attribute information based on characteristic information corresponding to users of interest to the given user.

[0090] [Variation 1] In the above example, we described an example in which posting information and user information, which are examples of target information, are managed by the service provider server 200. However, this is not limited to this example, and posting information and user information may be managed by other devices. For example, there may be a management server that can communicate with the service provider server 200 and the information processing device 100. The management server may manage the posting information and user information. In this case, the management server obtains the posting information and user information from the service provider server 200 as needed. The management server then stores the obtained information. The information processing device 100 extracts the information to be used as target information from the posting information and user information managed by the management server.

[0091] Furthermore, the information processing device 100 may manage the posted information and user information. In this case, the information processing device 100 acquires the posted information and user information from the service provision server 200 as needed. The information processing device 100 then stores the acquired information in the storage device 190. The information processing device 100 extracts the target information from the posted information and user information stored in the storage device 190.

[0092] <Third Embodiment> Next, an information processing device of the third embodiment will be described. Posted information, which is an example of target information, may have attribute information assigned to it in advance. In the third embodiment, an example of a method for assigning attribute information to posted information will be described. Note that some explanations will be omitted as they overlap with the first and second embodiments.

[0093] The information processing device 101 in this embodiment can be configured in the same way as the information processing device 100 shown in Figure 1. That is, the information processing device 101 is connected to the service provision server 200 and the user terminal 300 in a communicative manner.

[0094] Furthermore, the information processing device 101 is capable of performing the same processing as the information processing device 100. In addition, the information processing device 101 can perform the processing described below.

[0095] Figure 8 is a block diagram showing an example of the functional configuration of the information processing device 101. The information processing device 101 comprises an extraction unit 110, a identification unit 120, a generation unit 130, and an output unit 140. The information processing device 101 may also comprise a determination unit 150, an association unit 160, and a search unit 170. Furthermore, the information processing device 101 may comprise a storage device 190. The storage device 190 may be a device owned by the information processing device 101, or it may be an external device that is communicatively connected to the information processing device 101.

[0096] The information processing device 101 assigns attribute information to the posted information. The attribute information assigned to the posted information is treated as characteristic information when generating attribute information for a predetermined user. Hereafter, the posted information to which attribute information is assigned will also be referred to as the target posted information. In this case, each piece of posted information is assumed to include an image.

[0097] The decision unit 150 determines the target post information. For example, the decision unit 150 determines that post information that does not have attribute information attached is the target post information. Alternatively, the decision unit 150 may determine that post information that has had attribute information attached for a predetermined period of time is the target post information.

[0098] The search unit 170 performs a search. Specifically, the search unit 170 searches for similar images that are similar to the images contained in the target post information. In this case, the search unit 170 searches for similar images, for example, from the post information managed by the service provision server 200. The search unit 170 is not limited to this example, and may also search for similar images from post information in other online services managed by servers that provide other online services.

[0099] For example, the search unit 170 compares images included in target posts with images included in posts to be searched and calculates similarity. For example, the search unit 170 extracts features related to objects, edges, and patterns in the images using feature extraction algorithms such as SIFT (Scale Invariant Feature Transform) and SURF (Speeded-Up Robust Features). The search unit 170 may then calculate the similarity by calculating the distance between the features of the images. Note that the method for calculating similarity is not limited to this example. Similarity may be calculated using various known techniques.

[0100] Assuming that images are more similar the higher their similarity score, the search unit 170 may output images with a similarity score higher than the threshold as search results. In other words, the posted information that the search unit 170 finds will include similar images with a similarity score of 1 or higher than the threshold.

[0101] Furthermore, the search unit 170 may perform searches using other methods. For example, the search unit 170 may divide an image contained in the target post information into elements of an object and a background. An example of an object is a person. That is, the search unit 170 may divide an image contained in the target post information into an area of ​​the person and an area of ​​the background that is not the person. Another example of an object is an item worn by the person. Items include, for example, clothes, hats, bags, shoes, glasses, and accessories. That is, the search unit 170 may divide an image contained in the target post information into an area of ​​the item and an area of ​​the background that is not the item.

[0102] The search unit 170 may then search for similar images for each divided element. Specifically, the search unit 170 may search the posted information for images that contain elements similar to the divided elements. In this case, images that contain elements similar to the divided elements are considered similar images.

[0103] In this way, the search unit 170 searches for images. The search unit 170 is an example of a search means. Specifically, the search unit 170 searches for similar images that are similar to the images contained in the posted information (target posted information). For example, the search unit 170 may search for similar images that are contained in other posted information in an online service and are similar to the images contained in the posted information. Alternatively, for example, the search unit 170 may divide the image contained in the posted information into at least two elements: an object and a background, and then search for similar images that contain elements similar to the divided elements.

[0104] The identification unit 120 identifies attribute information attached to similar images. Specifically, the identification unit 120 identifies attribute information attached to other posted information that is found in the search performed by the search unit 170. The identification unit 120 may also identify attribute information attached to similar images contained in other posted information that is found in the search.

[0105] The generation unit 130 generates the identified attribute information as attribute information for the target post information. Specifically, the generation unit 130 may generate the identified attribute information itself as attribute information to be attached to the target post information. Also, if multiple pieces of attribute information are identified, the generation unit 130 may generate the attribute information for which the number of pieces of attribute information of the same type is equal to or greater than a threshold as attribute information for the target post information. The method for generating attribute information can be carried out in various ways. For example, the generation unit 130 may generate the attribute information attached to the post information containing the similar image with the highest similarity among the identified attribute information as attribute information for the target post information.

[0106] The generated attribute information corresponds to feature information that indicates characteristics related to the target post information. In other words, the generation unit 130 generates the identified attribute information as feature information that indicates characteristics related to the post information.

[0107] The association unit 160 associates the generated attribute information with the target post information. The post information is managed, for example, by the service provider server 200. The association unit 160 associates the generated attribute information with the target post information from among the post information managed by the service provider server 200 and registers it.

[0108] [Example of operation of the information processing device 101] Next, an example of the operation of the information processing device 101 will be explained using Figure 9. This example of operation will explain how to add feature information to posted information.

[0109] Figure 9 is a third flowchart illustrating an example of the operation of the information processing device 101. First, the determination unit 150 determines the target posting information (S201). For example, the determination unit 150 determines that posting information from online services that does not have characteristic information attached is the target posting information.

[0110] The search unit 170 searches for similar images that are similar to the image contained in the target post information (S202). For example, the search unit 170 searches for similar images in other posts on an online service that are similar to the image contained in the target post information.

[0111] The identification unit 120 identifies attribute information attached to similar images (S203). The generation unit 130 generates feature information for the target post information (S204). For example, the generation unit 130 generates the identified attribute information as feature information for the target post information. Then, the association unit 160 associates the generated feature information with the target post information (S205).

[0112] This example of operation is merely one example. In other words, the operation of the information processing device 101 in this disclosure is not limited to this example.

[0113] In this way, the information processing device 101 of the third embodiment searches for images. At this time, the posted information includes images. The information processing device 101 then searches for similar images that are similar to the images included in the posted information, identifies the attribute information attached to the similar images, and generates the identified attribute information as feature information that indicates features related to the posted information.

[0114] For example, the information processing device 101 may search for similar images in an online service that are similar to the images contained in other posted information. The information processing device 101 may then identify attribute information attached to the other posted information that matches the search, and generate characteristic information that indicates features related to the posted information from the identified attribute information.

[0115] As a result, the information processing device 101 can generate more appropriate characteristic information for the posted information.

[0116] Furthermore, the information processing device 101 may divide the image included in the posted information into at least two elements: an object and a background, and search for similar images that contain elements similar to the divided elements. Simply searching for images similar to the entire target image may not yield appropriate similar images, such as images that do not match the purpose of the target image. In other words, there may be accuracy issues when searching for similar images to be used as reference for generating attribute information to be attached to the posted information. In contrast, the information processing device 101 divides the image into elements and searches for images that contain elements similar to the divided elements, so it can find similar images that better match the purpose of the target image.

[0117] <Examples of Hardware Configuration of Information Processing Devices> The hardware constituting the information processing devices of the first, second, and third embodiments described above will now be explained. Figure 10 is a block diagram showing an example of the hardware configuration of the computer device constituting the information processing device in each embodiment. The computer device 90 realizes the information processing device and information processing method described in each embodiment and each modified example. For example, the information processing device etc. described in each embodiment and each modified example may have the hardware configuration shown in Figure 10.

[0118] As shown in Figure 10, the computer device 90 includes a processor 91, RAM (Random Access Memory) 92, ROM (Read Only Memory) 93, storage device 94, input / output interface 95, bus 96, and drive device 97. Note that the information processing device and the like may be implemented by multiple electrical circuits.

[0119] The storage device 94 stores a program (computer program) 98. The processor 91 executes the program 98 of this information processing device using the RAM 92. Specifically, for example, the program 98 includes a program that causes a computer to execute the processes shown in Figures 3, 7, and 9. The functions of each component of this information processing device are realized in response to the processor 91 executing the program 98. The program 98 may also be stored in the ROM 93. Alternatively, the program 98 may be recorded on the recording medium 80 and read using the drive device 97, or it may be transmitted to the computer device 90 from an external device (not shown) via a network (not shown).

[0120] The input / output interface 95 exchanges data with peripheral devices (keyboard, mouse, display device, etc.) 99. The input / output interface 95 functions as a means of acquiring or outputting data. The bus 96 connects each component.

[0121] Furthermore, there are various variations in how information processing devices are implemented. For example, each component included in an information processing device can be implemented as a dedicated device. Also, each information processing device can be implemented based on a combination of multiple devices.

[0122] The processing method for recording a program to realize each configuration in the function of each embodiment on a recording medium, reading the program recorded on the recording medium as code, and executing it on a computer is also included in the scope of each embodiment. In other words, a computer-readable recording medium is also included in the scope of each embodiment. Furthermore, the recording medium on which the above-mentioned program is recorded, and the program itself, are also included in each embodiment.

[0123] The recording medium in question is, but is not limited to, a floppy disk, hard disk, optical disk, magneto-optical disk, CD (Compact Disc)-ROM, magnetic tape, non-volatile memory card, or ROM. Furthermore, the programs recorded on the recording medium are not limited to programs that perform processing independently, but also include programs that operate on the OS (Operating System) in cooperation with other software and the functions of expansion boards to perform processing, and these are also included in the scope of each embodiment.

[0124] Although the present invention has been described above with reference to embodiments, the present invention is not limited to the above embodiments. Various modifications to the structure and details of the present invention can be made within the scope of the present invention as can be understood by those skilled in the art.

[0125] Furthermore, the above embodiments and modifications can be combined as appropriate.

[0126] The above embodiments may also be described in part or in whole as follows, but are not limited to these.

[0127] <Note> [Note 1] An information processing device comprising: an extraction means for extracting target information which is information relating to the target of an action by a predetermined user in an online service where information is posted; an identification means for identifying feature information which indicates features related to the extracted target information; an output means for generating attribute information which indicates the attributes preferred by the predetermined user based on the identified feature information; and an output means for outputting basis information which is information relating to the actions of the predetermined user and indicates the basis for the generation of attribute information for the predetermined user.

[0128] [Note 2] The information processing apparatus according to Note 1, wherein the target information includes posted information indicating information posted on the online service, the extraction means extracts the posted information to which a predetermined user has reacted, and the identification means identifies the feature information indicating features related to the extracted posted information.

[0129] [Note 3] The information processing apparatus according to Note 2, wherein the posted information includes at least one of an image and text, and the identifying means identifies at least one of the following as feature information: attribute information attached to the extracted posted information and keywords related to the content of the posted information, which are estimated from at least one of the image and text included in the posted information.

[0130] [Note 4] The information processing apparatus according to any one of Notes 1 to 3, wherein the target information includes user information which is information relating to a user in the online service, the extraction means extracts user information relating to other users that have a relationship with the predetermined user, and the identification means identifies attribute information attached to the extracted user information as characteristic information.

[0131] [Note 5] The information processing apparatus according to any one of Notes 1 to 4, wherein the identifying means identifies a plurality of feature information, and the generating means generates, among the plurality of identified feature information, the feature information in which the number of feature information of the same type is equal to or greater than a threshold, as attribute information for the predetermined user.

[0132] [Appendix 6] The information processing apparatus according to any one of Appendix 1 to 4, wherein the identified feature information is weighted according to the properties of the target information corresponding to the feature information, and the generation means generates attribute information for the predetermined user according to the weights of the identified feature information.

[0133] [Note 7] The information processing apparatus according to any one of Notes 1 to 6, wherein the output means outputs attribute information for the predetermined user and basis information for each attribute information, and the basis information includes information indicating which action of the predetermined user the characteristic information corresponding to the generated attribute information is based on.

[0134] [Appendix 8] An information processing device according to Appendix 2 or 3, comprising a search means for searching for images, wherein the posted information includes images, the search means searches for similar images similar to the images included in the posted information, the identification means identifies attribute information attached to the similar images, and the generation means generates the identified attribute information as feature information indicating features related to the posted information.

[0135] [Note 9] The information processing apparatus according to Note 8, wherein the search means searches for similar images in the online service that are similar to the images in the posted information, the identification means identifies attribute information attached to the other posted information that is found in the search, and the generation means generates the identified attribute information as feature information that indicates features related to the posted information.

[0136] [Note 10] The information processing apparatus according to Note 8 or 9, wherein the search means divides the image contained in the posted information into at least the elements of an object and a background, and searches for similar images that contain elements similar to the divided elements.

[0137] [Note 11] If the target information includes posted information indicating information posted on the online service, the extraction means extracts the posted information to which the predetermined user has reacted, the identification means identifies the feature information indicating features related to the extracted posted information, and the identified feature information is weighted according to at least one of the type of posted information corresponding to the feature information and the poster of the posted information, as described in Note 6.

[0138] [Note 12] If the target information includes user information which is information about a user in the online service, the extraction means extracts user information relating to other users who have a relationship with the predetermined user, the identification means identifies attribute information attached to the extracted user information as feature information, and the identified feature information is weighted according to the predetermined user's reaction to the other user's posted information corresponding to the feature information, as described in Note 6.

[0139] [Note 13] The information processing apparatus according to any one of Notes 1 to 12, wherein the attribute information includes information indicating human sensibilities toward an object.

[0140] [Note 14] An information processing method that extracts target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted; identifies feature information that shows characteristics related to the extracted target information; generates attribute information that shows attributes preferred by the predetermined user based on the identified feature information; and outputs basis information that shows the basis for the generation of attribute information for the predetermined user, relating to the actions of the predetermined user.

[0141] [Note 15] A recording medium that stores a program that causes a computer to execute the following: a process for extracting target information which is information relating to the target of an action by a predetermined user in an online service where information is posted; a process for identifying feature information which indicates features related to the extracted target information; a process for generating attribute information which indicates the attributes preferred by the predetermined user based on the identified feature information; and a process for outputting basis information which is information relating to the actions of the predetermined user and indicates the basis for the generation of attribute information for the predetermined user.

[0142] Furthermore, some or all of the configurations described in Appendices 2 to 13, which are dependent on Appendice 1 above, may also be dependent on Appendices 14 and 15 in the same way as Appendices 2 to 13. Moreover, within the scope that does not depart from each of the embodiments described above, some or all of the configurations described as appendices may also be dependent on various hardware, software, various recording means for recording software, or systems.

[0143] 100, 101 Information processing device 110 Extraction unit 120 Identification unit 130 Generation unit 140 Output unit 150 Determination unit 160 Association unit 170 Search unit 190 Storage device 200 Service provision server 300 User terminal 1000 Information processing system

Claims

1. An information processing device comprising: an extraction means for extracting target information which is information relating to the target of an action by a predetermined user in an online service where information is posted; an identification means for identifying feature information which indicates features related to the extracted target information; a generation means for generating attribute information which indicates the attributes preferred by the predetermined user based on the identified feature information; and an output means for outputting basis information which is information relating to the actions of the predetermined user and indicates the basis for the generation of attribute information for the predetermined user.

2. The information processing apparatus according to claim 1, wherein the target information includes posted information indicating information posted on the online service, the extraction means extracts the posted information to which a predetermined user has reacted, and the identification means identifies the feature information indicating features related to the extracted posted information.

3. The information processing apparatus according to claim 2, wherein the posted information includes at least one of an image and text, and the identifying means identifies at least one of attribute information attached to the extracted posted information and keywords related to the content of the posted information, which are estimated from at least one of the image and text included in the posted information, as the feature information.

4. The information processing apparatus according to any one of claims 1 to 3, wherein the target information includes user information which is information relating to a user in the online service, the extraction means extracts user information relating to other users which have a relationship with the predetermined user, and the identification means identifies attribute information attached to the extracted user information as characteristic information.

5. The information processing apparatus according to any one of claims 1 to 4, wherein the identifying means identifies a plurality of feature information, and the generating means generates, among the plurality of identified feature information, the feature information in which the number of feature information of the same type is equal to or greater than a threshold, as attribute information for the predetermined user.

6. The information processing apparatus according to any one of claims 1 to 4, wherein the identified feature information is weighted according to the properties of the target information corresponding to the feature information, and the generation means generates attribute information for the predetermined user according to the weights of the identified feature information.

7. The information processing apparatus according to any one of claims 1 to 6, wherein the output means outputs attribute information for the predetermined user and basis information for each attribute information, and the basis information includes information indicating which action of the predetermined user the characteristic information corresponding to the generated attribute information is based on.

8. An information processing apparatus according to claim 2 or 3, comprising a search means for searching for images, wherein the posted information includes images, the search means searches for similar images similar to the images included in the posted information, the identification means identifies attribute information attached to the similar images, and the generation means generates the identified attribute information as feature information indicating features related to the posted information.

9. The information processing apparatus according to claim 8, wherein the search means searches for similar images in the online service that are similar to the images in the posted information, the identification means identifies attribute information attached to the other posted information that is found in the search, and the generation means generates the identified attribute information as feature information indicating features related to the posted information.

10. The information processing apparatus according to claim 8 or 9, wherein the search means divides the image contained in the posted information into at least the elements of an object and a background, and searches for similar images that include elements similar to the divided elements.

11. If the target information includes posted information indicating information posted on the online service, the extraction means extracts the posted information to which a predetermined user has reacted; the identification means identifies the feature information indicating features related to the extracted posted information; and the identified feature information is weighted according to at least one of the type of posted information corresponding to the feature information and the poster of the posted information, as described in claim 6.

12. If the target information includes user information which is information relating to a user in the online service, the extraction means extracts user information relating to other users who have a relationship with the predetermined user; the identification means identifies attribute information attached to the extracted user information as feature information; and the identified feature information is weighted according to the predetermined user's reaction to the other user's posted information corresponding to the feature information, as described in claim 6.

13. The information processing apparatus according to any one of claims 1 to 12, wherein the attribute information includes information indicating human sensibilities toward an object.

14. An information processing method comprising: extracting target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted; identifying characteristic information that shows features related to the extracted target information; generating attribute information that shows attributes preferred by the predetermined user based on the identified characteristic information; and outputting basis information that shows the basis for the generation of attribute information for the predetermined user, relating to the actions of the predetermined user.

15. A recording medium that stores a program that causes a computer to execute the following: a process for extracting target information, which is information relating to the target of an action by a predetermined user in an online service where information is posted; a process for identifying feature information that indicates characteristics related to the extracted target information; a process for generating attribute information that indicates attributes preferred by the predetermined user based on the identified feature information; and a process for outputting basis information that indicates the basis for generating attribute information for the predetermined user, which is information relating to the actions of the predetermined user.

Images