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

Abstract

To improve the convenience of a data distribution service, an information processing device (1) includes: a related user extraction unit (11) for extracting a related user related to a target user on the basis of target user associated information associated with the target user; and a recommended information generation unit (12) for generating recommended information indicating data that is recommended to be acquired by the target user on the basis of related user associated information associated with the related user.

Description

[Technical Field]

[0001] The present invention relates to an information processing device or the like that recommends data in a data distribution service. [Background technology]

[0002] In recent years, data distribution services that allow other users to acquire data registered by a user have become popular. Users of data distribution services register data with the data distribution service and acquire data registered by other users via the distribution service. While such data distribution services are convenient, because a wide variety of data is registered on data distribution services, users may find it difficult to find data that is useful to them on their own.

[0003] Prior art for solving such problems includes, for example, a data management device described in Patent Document 1 below. This data management device extracts data related to data provided by a service subscriber based on the metadata of the data, and recommends the extracted data to the subscriber. This configuration makes it easier for subscribers to find data that is useful to them. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-024645 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology described in Patent Document 1 recommends data related to data provided by the subscribers themselves, so the recommended data is limited to data within the range that the subscribers can predict. As such, the technology described in Patent Document 1 has room for improvement in terms of improving the convenience of data distribution services.

[0006] One aspect of the present invention has been made in consideration of the above-mentioned problems, and one example of its purpose is to provide an information processing device or the like that can improve the convenience of data distribution services. [Means for solving the problem]

[0007] An information processing device according to one aspect of the present invention comprises: a related user extraction means for extracting related users associated with a target user from among a plurality of users of a data distribution service that enables other users to acquire data registered by a given user, based on target user correspondence information associated with the target user; and a recommendation information generation means for generating recommendation information indicating data that is recommended for the target user to acquire from among the data that can be acquired through the data distribution service, based on the related user correspondence information associated with the related user.

[0008] A recommended information generation method according to one aspect of the present invention includes at least one processor extracting related users associated with a target user from among multiple users of a data distribution service that enables other users to acquire data registered by a user, based on target user correspondence information associated with the target user, and generating recommended information indicating data that is recommended for the target user to acquire from among the data that can be acquired through the data distribution service, based on the related user correspondence information associated with the related users.

[0009] A recommended information generation program according to one aspect of the present invention causes a computer to function as a related user extraction means that extracts related users related to a target user from among multiple users of a data distribution service that enables other users to acquire data registered by a user, based on target user correspondence information associated with the target user, and a recommended information generation means that generates recommended information indicating data that is recommended for the target user to acquire from among the data that can be acquired through the data distribution service, based on the related user correspondence information associated with the related user. [Effects of the Invention]

[0010] According to one aspect of the present invention, the convenience of data distribution services can be improved. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to a first exemplary embodiment of the present invention. [Figure 2] 1 is a flowchart showing the flow of a recommendation information generating method according to a first exemplary embodiment of the present invention. [Figure 3] FIG. 10 is a diagram illustrating an overview of a recommendation information generation system according to a second exemplary embodiment of the present invention. [Figure 4] FIG. 2 is a block diagram showing an example of a configuration of a main part of an information processing device included in the recommendation information generation system. [Figure 5] FIG. 2 is a diagram illustrating an example of management information stored in the information processing device. [Figure 6] FIG. 10 is a flowchart showing the flow of a recommendation information generating method executed by the information processing device. [Figure 7] FIG. 10 is a diagram illustrating a specific example of the recommendation information generating method. [Figure 8] FIG. 10 is a diagram showing an example of generating recommended information based on the frequency of appearance in interest information and the frequency of appearance in FB information. [Figure 9] FIG. 10 is a diagram illustrating an example of generating recommendation information based on the frequency of appearance of each business type. [Figure 10] FIG. 10 is a diagram illustrating an example of generation of recommendation information based on a user's specification. [Figure 11] FIG. 10 is a flowchart showing the flow of a method for generating recommended information using preference information. [Figure 12] FIG. 1 is a diagram illustrating an example of a computer that executes instructions of a program that is software that realizes each function of an information processing device according to each exemplary embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] Exemplary Embodiment 1 A first exemplary embodiment of the present invention will be described in detail with reference to the drawings. This exemplary embodiment is a basic form of the exemplary embodiments described below.

[0013] (Configuration of information processing device) The configuration of an information processing device 1 according to this exemplary embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in the figure, the information processing device 1 includes a related user extraction unit 11 and a recommended information generation unit 12.

[0014] The related user extraction unit 11 extracts related users related to a target user from among multiple users of a data distribution service that enables other users to acquire data registered by a user (hereinafter referred to as registered data), based on target user correspondence information associated with the target user.

[0015] The recommended information generation unit 12 generates recommended information indicating registration data (hereinafter referred to as recommended registration data) that is recommended for the target user to acquire from among the registration data that can be acquired through the data distribution service, based on the related user correspondence information associated with the related user.

[0016] As described above, the information processing device 1 according to this exemplary embodiment is configured to include a related user extraction unit 11 that extracts related users associated with a target user from among multiple users of a data distribution service based on target user association information associated with the target user, and a recommended information generation unit 12 that generates recommended information indicating recommended registration data based on the related user association information associated with the related users. Therefore, the information processing device 1 according to this exemplary embodiment can recommend registration data that the target user would find difficult to recognize as useful to the target user, thereby improving the convenience of the data distribution service.

[0017] (Recommendation Generator) The functions of the information processing device 1 described above can also be realized by a program. A recommended information generation program according to this exemplary embodiment is configured to cause a computer to function as: a related user extraction unit that extracts related users associated with a target user from among multiple users of a data distribution service based on target user association information associated with the target user; and a recommended information generation unit that generates recommended registration data that is recommended for the target user to acquire from among the registration data available through the data distribution service based on the related user association information associated with the related users. Therefore, the recommended information generation program according to this exemplary embodiment can recommend registration data that the target user would find difficult to recognize as useful to the target user, thereby improving the convenience of the data distribution service.

[0018] (Flow of recommendation information generation method) The flow of the recommendation information generation method according to this exemplary embodiment will be described with reference to Fig. 2. Fig. 2 is a flow diagram showing the flow of the recommendation information generation method. Note that the execution entity of each step in this recommendation information generation method may be a processor provided in the information processing device 1, or a processor provided in another device, or each step may be executed by a processor provided in a different device.

[0019] In S11, at least one processor extracts related users associated with a target user from among a plurality of users of the data distribution service, based on target user correspondence information associated with the target user.

[0020] In S12, at least one processor generates recommendation information indicating recommended registration data that is recommended for the target user to acquire from among the registration data that can be acquired through the data distribution service, based on related user correspondence information associated with the related users extracted in S11.

[0021] As described above, the recommendation information generation method according to this exemplary embodiment is configured to extract related users associated with a target user from among multiple users of a data distribution service based on target user association information associated with the target user, and to generate recommendation information indicating recommended registration data that is recommended for the target user to acquire from among the registration data available through the data distribution service based on the related user association information associated with the related users. Therefore, the recommendation information generation method according to this exemplary embodiment can recommend to the target user registration data that the target user would find difficult to recognize as useful, thereby achieving the effect of improving the convenience of the data distribution service.

[0022] Exemplary Embodiment 2 (Outline of recommendation information generation system 100) A second exemplary embodiment of the present invention will be described in detail with reference to Fig. 3 to Fig. 11. First, an overview of a recommendation information generation system 100 according to this embodiment will be described with reference to Fig. 3. Fig. 3 is a diagram showing an overview of the recommendation information generation system 100.

[0023] As shown in the figure, the recommendation information generation system 100 includes an information processing device 2 and a terminal device 3. The information processing device 2 is a device that provides a platform for the data distribution service, and the terminal device 3 is a device used by a target user who uses the data distribution service.

[0024] A data distribution service is a service that allows other users to obtain data registered by a user. Below, an example will be described in which the data distribution service is a registration-based service that can only be used by registered users, and requires users to register data that they are willing to provide to other users as a registration requirement. Also, here, an example will be described in which the user of the data distribution service is a company. Of course, the data distribution service may also be available to unregistered users or users who do not provide data, and the users of the data distribution service are not limited to companies.

[0025] The registered data may be any data that is useful to users of the data distribution service, but it is preferable that the data be data that each user has independently collected. For example, if a user is a store owner, the registered data may be a data table showing the sales trends and customer numbers of the store that the user manages.

[0026] The information processing device 2 accepts registrations for use of the data distribution service. As described above, registered users are required to register data, so the information processing device 2 accepts uploads of registered data. Furthermore, the information processing device 2 also accepts requests to acquire registered data and performs processing to download the registered data to the requesting user.

[0027] For example, in the example in Figure 3, a total of five companies, Company A to Company E, are registered with the data distribution service. All of these companies provide registered data to the data distribution service, and each company can obtain registered data provided by other companies.

[0028] In addition, each of these companies also registers in the information processing device 2 company information indicating the company name, industry, etc., data summary information indicating an overview of the registered data provided by the company, and interest information indicating the registered data in which the company is interested in the data distribution service (hereinafter referred to as data of interest).

[0029] Furthermore, companies that have acquired registered data through the data distribution service provide feedback on the acquired registered data, and the acquired data and the content of the feedback are registered in the information processing device 2 as FB information.

[0030] The information processing device 2 also performs processing to present recommended registration data, which is registration data that is recommended to users of the data distribution service. Specifically, the information processing device 2 generates recommended information indicating the recommended registration data and displays this on the terminal device 3, thereby presenting the recommended registration data to the target user. For example, the information processing device 2 may display the recommended information on My Page, which serves as an interface when the user uses the data distribution service, or may notify the user of the recommended information by email or the like.

[0031] 3 shows an example in which recommended information is generated and presented to a target user, the target user being Company A, one of the companies registered with the data distribution service. In this case, the information processing device 2 first checks the information registered about Company A, specifically, the company information and data summary information of Company A (S1).

[0032] Next, based on the information confirmed in S1, the information processing device 2 extracts related users related to company A from among the users registered in the data distribution service (S2). In the example of Fig. 3, companies D and E are extracted.

[0033] Details will be given later, but in S2, based on various information about the target user (for example, information indicating the target user's industry and type of registered data) and various information about other users (for example, information indicating the industry and type of registered data), companies with industry and registered data similar to that of the target user can be extracted as related users.

[0034] Next, the information processing device 2 generates recommended information for the target user, Company A, based on related user association information associated with the related users extracted in S2, specifically, the interest information and Facebook information of Company D and Company E (S3).

[0035] In S3, the information processing device 2 may analyze the usage history and usage trends of the registered data of related users in the data distribution service, and generate recommendation information based on the analysis results. Note that the method of generating recommendation information will be described in detail later.

[0036] Then, the information processing device 2 presents the recommended information generated in S3 to company A (S4). The method of presenting the recommended information is not particularly limited, and for example, as described above, the recommended information may be displayed on the personal page that company A views on the terminal device 3.

[0037] (Configuration of information processing device 2) The configuration of the information processing device 2 will be described with reference to Fig. 4. Fig. 4 is a block diagram showing an example of the configuration of the main parts of the information processing device 2. As shown in the figure, the information processing device 2 includes a control unit 20 that controls all parts of the information processing device 2, and a storage unit 21 that stores various data used by the information processing device 2. The information processing device 2 also includes a communication unit 22 that enables the information processing device 2 to communicate with other devices, an input unit 23 that accepts input of various data to the information processing device 2, and an output unit 24 that enables the information processing device 2 to output various data.

[0038] The control unit 20 also includes a data management unit 201, a related user extraction unit 202, a candidate data extraction unit 203, a candidate data evaluation unit 204, and a recommended information generation unit 205. The storage unit 21 stores management information 211 and recommended information 212.

[0039] The data management unit 201 performs various processes necessary to provide the data distribution service. Specifically, the data management unit 201 manages users of the data distribution service, registered registration data, and various information related to these. For example, the data management unit 201 accepts user registrations for the data distribution service, accepts uploads of registered data, accepts requests to obtain registered data, and controls downloads.

[0040] The related user extraction unit 202 extracts related users related to a target user of the data distribution service from among users of the data distribution service, based on target user correspondence information associated with the target user. Details of the target user correspondence information and the method for extracting related users will be described later.

[0041] The candidate data extraction unit 203 extracts candidate data that is a candidate for recommended registration data that is recommended for acquisition by the target user from the registration data registered in the data distribution service. As will be described in detail later, the extraction of candidate data is performed based on the related user correspondence information of the related users extracted by the related user extraction unit 202.

[0042] The candidate data evaluation unit 204 evaluates each candidate data extracted by the candidate data extraction unit 203. This evaluation should be such that the more likely the candidate data is to be useful to the target user, the higher the evaluation. For example, candidate data that is commonly included in the related user association information of many related users is likely to be useful to the target user. Therefore, the candidate data evaluation unit 204 may calculate, for example, the frequency expressed by the following formula as the evaluation value.

[0043] (Frequency) = 100 × (number of candidate data included in related user correspondence information) / (number of extracted related users) The recommended information generating unit 205 generates recommended information indicating recommended registration data that is recommended for the target user to acquire from among the registration data that can be acquired through the data distribution service, based on the related user association information associated with the related user. Specifically, the recommended information generating unit 205 determines which candidate data is recommended for the target user based on the evaluation results for each candidate data extracted based on the related user association information, and generates recommended information that sets the candidate data as recommended registration data.

[0044] Management information 211 is information for managing users, registered data, and various information related to these in the data distribution service. Management information 211 may include the above-mentioned target user association information and related user association information. Details of management information 211 will be described later with reference to FIG. 5.

[0045] The recommended information 212 is information indicating recommended registration data that is recommended for acquisition by users of the data distribution service, and the recommended information 212 is generated by the recommended information generating unit 205 according to each user as described above.

[0046] (Details of Management Information 211) The management information 211 will be described in detail with reference to Fig. 5. Fig. 5 is a diagram showing an example of the management information 211. The management information 211 shown in the figure has a data structure in which a user ID, company name, industry, registered data, data summary, interest information, and Facebook information are associated with each other.

[0047] The user ID is identification information for uniquely identifying a user of the data distribution service. The company name is information indicating the name of the company to which the user belongs. The industry is information indicating the industry of the company to which the user belongs. The company name and industry are examples of company information shown in Figure 3. The company information may be any information about the company to which the user belongs, and may also include, for example, the location of the company's headquarters and information indicating the size of the company (number of employees, sales, etc.). In the example of Figure 5, the "company name" fields for the first and second users are "Company A" and "Company B," respectively, and the "industry" fields for both of these users are "retail."

[0048] The registration data is information indicating the registration data of a user of the data distribution service. The management information 211 may record the name of the registration data, or may record the storage location of the registration data or the identification information (ID) of the registration data. Note that one user may register multiple pieces of registration data. In the example of Figure 5, the "Registered Data" items for "Company A" and "Company B" have the registration data names "aaaa" and "bbbb", respectively.

[0049] The data summary is information that indicates an overview of the registered data, and is the same information as the "data summary information" shown in Figure 3. When multiple registered data are registered, a data summary is recorded for each registered data. The data summary may indicate, for example, the type of registered data, and for example, the metadata of the registered data may be used as the data summary. In the example of Figure 5, the "data summary" item for "Company A" and "Company B" is both "drink purchase data." This shows that the registered data for "Company A" and "Company B" is data related to the purchase of drinks.

[0050] Interest information is information about registered data in which related users are interested. Interest information may also indicate the industry or business type in which related users are interested. In the example of Figure 5, the "interest information" item for "Company A" is "xxyy data from Retail Company X," and the "interest information" item for "Company B" is "xxxx data from Insurance Company E."

[0051] The registered data indicated in the interest information is data of interest that the related user is particularly interested in among the registered data registered in the data distribution service. The interest information only needs to include information for identifying the data of interest (for example, the data name, ID, storage location, etc. of the data of interest).

[0052] The FB information is information indicating the content of user feedback on the acquired registered data. The FB information is recorded for each acquired registered data. The FB information may indicate, for example, the evaluation of the registered data by related users (e.g., a classification of high or low evaluations, or an evaluation value that expresses the evaluation numerically, etc.). The FB information may also indicate, for example, the quality of the registered data (e.g., the presence or absence of garbage data, the accuracy and granularity of the data, etc.), or the affinity with the registered data of the target user (e.g., the similarity with the registered data of the target user). Note that when no registered data has been acquired, as in the case of Company A in Figure 5, no FB information is recorded. On the other hand, in the example of Figure 5, Company B has acquired registered data named "zzzz," and it is recorded that the company provided "high" feedback for this data.

[0053] (Flow of recommendation information generation method) The flow of the recommendation information generation method according to this exemplary embodiment will be described with reference to Fig. 6. Fig. 6 is a flow diagram showing the flow of the recommendation information generation method executed by the information processing device 2. Note that the following also describes Fig. 7, which shows a specific example of the recommendation information generation method.

[0054] In S21, the related user extraction unit 202 extracts related users for a target user to whom recommended information is to be presented, based on target user association information associated with the target user. Specifically, the related user extraction unit 202 refers to the "data summary" of the management information 211, which is an example of target user association information, to identify users whose summary information is the same as that of the target user, and extracts the users as related users.

[0055] For example, in the example of FIG. 7, the target user is Company A, and the registered data of Company A is "drink purchase data." The fact that the registered data of Company A is "drink purchase data" is recorded in the "data summary" of management information 211 (see FIG. 5). By referring to the "data summary" of management information 211, the related user extraction unit 202 extracts companies B and D, which, like Company A, have "drink purchase data" as registered data, as related users.

[0056] The related user extraction unit 202 also extracts Company C, whose registered data is in the broader category of "food and beverage purchasing data" that includes "beverage purchasing data," as a related user. In this way, in S21 of FIG. 6, the related user extraction unit 202 may extract users who have registered data in the same category as the registered data of the target user, or may extract users who have registered data in a category corresponding to the registered data of the target user. It is only necessary to determine in advance which categories are to be regarded as corresponding categories. For example, categories containing the same character string, such as "beverage purchasing data" and "food and beverage purchasing data," may be regarded as corresponding categories.

[0057] In S22, the candidate data extraction unit 203 extracts candidate data based on the related user association information of the related user extracted in S21. Specifically, the candidate data extraction unit 203 extracts, as candidate data, the "acquired data" indicated in the "FB information" of the related user recorded in the management information 211. In addition, the candidate data extraction unit 203 extracts, as candidate data, data of interest indicated in the "interest information" of the related user recorded in the management information 211.

[0058] 7, the interest information of Company B, one of the extracted related users, is "xxxx data of Insurance Company E," so the candidate data extraction unit 203 extracts "xxxx data" as candidate data. Similarly, the candidate data extraction unit 203 extracts "xxxx data" as candidate data from the interest information of Company C, and extracts "yyyy data" as candidate data from the interest information of Company D.

[0059] Furthermore, the candidate data extraction unit 203 extracts the "zzzz data," "yyyy data," and "xxxx data" recorded as "acquired data" in the "FB information" of companies B to D as candidate data. Note that candidate data that has been extracted multiple times, such as "xxxx data," is merged. As a result, in the example of FIG. 7, three types of data, "xxxx data," "yyyy data," and "zzzz data," are extracted as candidate data.

[0060] In S22, it is not necessary to extract candidate data from both the interest information and the FB information; for example, it may be extracted from the interest information but not from the FB information, or it may be extracted from the FB information but not from the interest information.

[0061] In S23, the candidate data evaluation unit 204 calculates an evaluation value for each candidate data extracted in S22, both by industry type and regardless of industry. Specifically, the candidate data evaluation unit 204 calculates an evaluation value for all candidate data extracted in S22, and also calculates an evaluation value for candidate data extracted from interest information or Facebook information of related users in the same industry as the target user, among the candidate data extracted in S22.

[0062] For example, in the example of Fig. 7, the frequency that the candidate data is included in the related user information of the related users extracted in S21, i.e., the interest information and Facebook information, is calculated as the evaluation value. The calculation formula for the frequency that applies to all candidate data is as described above.

[0063] Specifically, one of the candidate data, "xxxx data," has a frequency of 100% because it is included in the "interest information" of companies B and C, and in the "FB information" of company D. The remaining candidate data, "Railway company F's yyyy data" and "zzzz data," have frequencies of 67% and 33%, respectively.

[0064] On the other hand, the frequency of occurrence by industry can be calculated using the following formula.

[0065] (Frequency by industry) = (number of candidate data included in the related user correspondence information of related users in a specific industry) / (number of extracted related users in a specific industry) In the example of Figure 7, the frequency is calculated for target users in the retail industry and related users in the same industry, namely, retail companies B and C. Specifically, one of the candidate data, "xxxx data," is included in the "interest information" of companies B and C, so the frequency is 100%. The remaining candidate data, "yyyy data" and "zzzz data," each have a frequency of 50%.

[0066] In S24, the recommendation information generation unit 205 determines whether or not to generate recommendation information based on the calculation result of S23. For example, the recommendation information generation unit 205 may determine to generate recommendation information if an evaluation value equal to or greater than a predetermined threshold is calculated in S23, and may determine not to generate recommendation information if an evaluation value equal to or greater than the predetermined threshold is not calculated. If the determination in S24 is YES, the process proceeds to S25, and if the determination in S24 is NO, the recommendation information generation method ends.

[0067] For example, in the example of Fig. 7, if a determination is made as to whether or not to generate recommended information using 60% as a threshold, then "xxxx data" with a frequency of 100% exists for the same industry, and therefore a YES determination is made in S24. Similarly, "xxxx data" also has a frequency of 100% for all industries, and therefore a YES determination is made in S24. Note that if it is determined that recommended information should be generated for at least either the same industry or all industries, the determination result in S24 will be YES.

[0068] In S25, the recommended information generation unit 205 generates recommended information based on the calculation result of S23. As described above, the evaluation value (frequency in the example of FIG. 7) is calculated based on related user associated information (interest information and Facebook information) associated with the related user, so in S25, recommended information is generated based on related user associated information associated with the related user. Then, the recommended information generation unit 205 stores the generated recommended information in the storage unit 21 as recommended information 212. This completes the recommended information generation method.

[0069] For example, in the example of Fig. 7, recommendation information is generated and stored that recommends the acquisition of "xxxx data" from Insurance Company E for both the same industry and all industries. Note that if there are multiple candidate data whose frequency is equal to or exceeds a threshold, recommendation information may be generated for each of those candidate data.

[0070] In addition to the registered data name, the recommended information may also include the registered user of the registered data. For example, in the example of Fig. 7, the recommended information generating unit 205 may generate recommended information indicating "xxxx data of Insurance Company E".

[0071] (Regarding extraction of related users) As described above, the target user association information may include information indicating the category of registered data that the target user has registered in the data distribution service. In this case, in the information processing device 2 according to this exemplary embodiment, the related user extraction unit 202 may extract, as related users, users who have registered in the data distribution service data of the same or corresponding category as the data that the target user has registered in the data distribution service.

[0072] A user who has registered data in the same or corresponding category as the target user can be said to have similar characteristics to the target user, and therefore registered data that is useful to such a user is likely to be useful to the target user as well. Therefore, the information processing device 2 according to this exemplary embodiment can achieve the effect of increasing the possibility of generating recommended information that indicates registered data that is useful to the target user, in addition to the effect achieved by the information processing device 1 according to exemplary embodiment 1.

[0073] The target user correspondence information may also include information indicating the business type of the target user. In this case, in the information processing device 2 according to this exemplary embodiment, the related user extraction unit 202 may extract users in the same or corresponding business type as the target user as related users in S21 of Fig. 6. It is only necessary to determine in advance which business types are to be considered as corresponding business types.

[0074] Users in the same or corresponding industry as the target user can be said to have similar characteristics to the target user, and therefore registered data useful to such users is likely to be useful to the target user as well. Therefore, the information processing device 2 according to this exemplary embodiment can achieve the effect of increasing the possibility of generating recommended information indicating registered data useful to the target user, in addition to the effect achieved by the information processing device 1 according to exemplary embodiment 1.

[0075] (About extracting related users using AI) The related user extraction unit 202 may extract related users using AI (Artificial Intelligence). For example, the related user extraction unit 202 may use AI that classifies registered data into categories to classify the registered data of the target user and the registered data of other users into respective categories. In this case, the related user extraction unit 202 may extract, as related users, users who have registered registered data that is classified into the same category as the registered data of the target user.

[0076] Such AI can be constructed by machine learning using registered data with known categories, and can also be called a classification model for category classification. Furthermore, various information related to the category of the registered data, such as metadata of the registered data, may be used in addition to or instead of the registered data for AI training.

[0077] Here, suppose the registered data is data indicating numerical values ​​for each series, such as a data table, but the series name is not indicated. It is generally difficult to classify such registered data into the correct category. Therefore, in such cases, the related user extraction unit 202 may use AI to estimate the series name from the distribution trend of the numerical values ​​of a certain series indicated in the registered data, and then perform category classification. In this case, the AI ​​can be constructed by machine learning using a group of numbers whose series names are known. Furthermore, if the registered data contains multiple series, and only some of the series names are not indicated, the series names not indicated in the registered data may be estimated taking into account the series names indicated in the registered data.

[0078] The related user extraction unit 202 may also extract related users by analyzing words included in the registered data of each user. For example, the related user extraction unit 202 may extract, as related users, users who have registered registered data (or metadata of registered data) that contains the same words or words with the same meaning as the registered data of the target user. For example, the related user extraction unit 202 may also calculate the similarity between registered data based on the number of the same words or words with the same meaning, and extract, as related users, users who have registered registered data with high similarity.

[0079] (About generating recommendations) As described above, the related user association information may include interest information indicating data of interest, which is data that the related user is interested in. In this case, in the information processing device 2 according to this exemplary embodiment, the recommendation information generation unit 205 may generate recommendation information that recommends acquiring the data of interest indicated in the interest information.

[0080] The data of interest is data in which related users related to the target user are interested, and therefore is likely to be useful to the target user. Therefore, in addition to the effects of the information processing device 1 according to the first exemplary embodiment, the information processing device 2 according to the present exemplary embodiment can also achieve an effect of increasing the possibility of generating recommended information indicating registered data that is useful to the target user.

[0081] As described above, the related user information may include information indicating acquired data that the related user has acquired through the data distribution service. In this case, in the information processing device 2 according to this exemplary embodiment, the recommendation information generation unit 205 may generate recommendation information that recommends acquiring the acquired data.

[0082] The acquired data actually acquired by related users related to the target user is likely to be useful to the target user. Therefore, the information processing device 2 according to this exemplary embodiment can achieve the effect of increasing the possibility of generating recommended information indicating registered data useful to the target user, in addition to the effect achieved by the information processing device 1 according to exemplary embodiment 1.

[0083] (Generating recommendation information taking into account Facebook information) As described above, the related user information may include feedback information indicating feedback content from related users regarding acquired data. In this case, in the information processing device 2 according to this exemplary embodiment, the recommendation information generator 205 may generate recommendation information that recommends acquiring acquired data for which feedback of predetermined content has been provided.

[0084] As described above, the acquired data actually acquired by related users is likely to be useful to the target user. The content of the feedback on the acquired data is a useful criterion for determining whether the acquired data is useful to the target user. Therefore, in addition to the effects of the information processing device 1 according to the first exemplary embodiment, the information processing device 2 according to this exemplary embodiment can further increase the possibility of generating recommended information indicating registered data useful to the target user.

[0085] For example, the recommendation information generating unit 205 may generate recommendation information that recommends acquiring acquired data that has received positive feedback. Conversely, the recommendation information generating unit 205 may generate recommendation information that recommends acquiring acquired data that has received negative feedback. This is because it is thought that there may be target users who are interested in registered data that other users have given negative feedback on.

[0086] Furthermore, for example, the recommendation information generation unit 205 may calculate the ratio of positive feedback to negative feedback and include the calculated ratio in the recommendation information. For example, if data X that appears most frequently in the FB information has 80% positive feedback and 20% negative feedback, the recommendation information generation unit 205 may designate data X as recommended registration data and may also include information indicating the ratio of the positive / negative feedback in the recommendation information.

[0087] (Example 1 of generating recommendation information) The recommended information may not only indicate the recommended registration data that is recommended for acquisition, but also indicate various analysis results regarding the recommended registration data. This will be explained with reference to Fig. 8. Fig. 8 is a diagram showing an example of generating recommended information based on the frequency of appearance in interest information and the frequency of appearance in Facebook information.

[0088] In the example of Fig. 8, it is assumed that the candidate data evaluation unit 204 calculates the frequency in the entire related user associated information including interest information and FB information as in the example of Fig. 7, and also calculates the frequency in interest information and the frequency in FB information. For example, in the example of Fig. 7, when calculating the frequency in interest information for all industries, the calculation formula is as follows:

[0089] (Frequency in interest information) = (number of candidate data included in interest information) / (number of extracted related users) For example, in the example of FIG. 7, when the frequency of interest information is calculated for all industries, the frequency of "xxxx data" is 67% and the frequency of "yyyy data" is 33%.

[0090] Furthermore, when calculating the frequency of appearance in interest information by industry, the calculation formula is as follows:

[0091] (Frequency in interest information of a specific industry) = (number of candidate data included in interest information of related users of a specific industry) / (number of extracted related users of a specific industry) Therefore, in the example of FIG. 7, when the frequency of appearance in interest information is calculated for related users in the same industry as the target user, the frequency of appearance of "xxxx data" is 100%.

[0092] The same is true for Facebook information: in the example of Figure 7, when the frequency of Facebook information is calculated for all industries, the frequency of "xxxx data," "yyyy data," and "zzzz data" is 33%. Also, in the example of Figure 7, when the frequency of Facebook information is calculated for related users in the same industry as the target user, the frequency of both "yyyy data" and "zzzz data" is 50%.

[0093] In this manner, candidate data evaluation unit 204 calculates six types of frequency for each piece of candidate data extracted by candidate data extraction unit 203. These six types of frequency are shown in a table at the top of Fig. 8.

[0094] As shown in this table, the six types of frequencies include three types for the same industry and three types for all industries. The former includes the frequency (a%) in the information of interest, the frequency (b%) in the FB information, and the frequency (c%) in the overall information of interest and FB information. Also, the latter also includes the frequency (a’%) in the information of interest, the frequency (b’%) in the FB information, and the frequency (c’%) in the overall information of interest and FB information.

[0095] The recommended information generation unit 205 generates recommended information based on the values of these six types of frequencies calculated for each candidate data. For example, the recommended information generation unit 205 may generate recommended information that recommends candidate data for which the frequency (c’%) in the overall information of interest and FB information for all industries is equal to or greater than the threshold value.

[0096] In this case, as shown by the frame line X1 in FIG. 8, the recommended information generation unit 205 may compare the frequency (a’%) in the information of interest for all industries with the frequency (c’%) in the overall information of interest and FB information for all industries. And when this comparison result is a’>c’, the recommended information generation unit 205 may generate recommended information indicating that the data named XXXX has gathered high interest across industries, as shown in FIG. 8. On the other hand, when a’<c’, the recommended information generation unit 205 may generate recommended information indicating that the data named XXXX has gathered a lot of feedback across industries. Note that the data named XXXX is one of the candidate data.

[0097] Similarly, when a>c, the recommended information generation unit 205 may generate recommended information indicating that the data named XXXX has gathered high interest within the same industry. Also, when a<c, the recommended information generation unit 205 may generate recommended information indicating that the data named XXXX has gathered a lot of feedback within the same industry.

[0098] In addition, when comparing a' and c', the recommendation information generator 205 may determine that a'>c' when the difference between a' and c' is equal to or greater than a threshold. For example, when the difference between a' and c' is 10% or greater, the recommendation information generator 205 may determine that a'>c'. Similarly, when the ratio between a' and c' is equal to or greater than a threshold, the recommendation information generator 205 may determine that a'>c'. For example, when the ratio between a' and c' is 1.2 or greater, the recommendation information generator 205 may determine that a'>c'. This also applies to the comparison between a and c described above and other conditions described below.

[0099] Furthermore, the recommendation information generation unit 205 may compare the frequency (a'%) of interest information targeting all industries with the frequency (b'%) of FB information targeting all industries, as shown by the frame line X2 in Fig. 8. If the comparison result is a'>b', the recommendation information generation unit 205 may generate recommendation information indicating that data XXXX is attracting interest across industries, but that there is little feedback on this data, as shown in Fig. 8.

[0100] Furthermore, the recommended information generating unit 205 may compare the frequency (a'%) of information of interest targeting all industries with the frequency (a%) of information of interest targeting the same industry, as shown by the frame line X3 in Fig. 8. If the comparison result is a'>a, the recommended information generating unit 205 may generate recommended information indicating that data XXXX is attracting interest in other industries, as shown in Fig. 8.

[0101] Furthermore, the recommendation information generation unit 205 may compare the overall frequency (c'%) of interest information and Facebook information targeting all industries with the overall frequency (c%) of interest information and Facebook information targeting the same industry, as shown by the frame line X4 in Fig. 8. If the comparison result is c'>c, the recommendation information generation unit 205 may generate recommendation information indicating that data XXXX is attracting interest or a lot of feedback in other industries, as shown in Fig. 8.

[0102] In addition, as shown by the frame line X5 in FIG. 8, the recommended information generation unit 205 may compare the frequency (b'%) in the FB information for all industries, the frequency (a%) of the interest information for the same industry, and the frequency (b%) of the FB information for the same industry. And when the comparison result is a'>b', as shown in FIG. 8, the recommended information generation unit 205 may generate recommended information indicating that the data of XXXX attracts interest across industries but has little feedback for this data.

[0103] In addition, the recommended information generation unit 205 may perform both the comparison shown by the frame line X1 and the comparison shown by the frame line X2 in FIG. 8. And when the comparison result is c'>c and a'<b', as shown in FIG. 8, the recommended information generation unit 205 may generate recommended information indicating that the data of XXXX attracts interest or a lot of feedback in other industries, and it can be seen that the amount of feedback is more prominent than the level of interest in other industries and the data is already in actual use.

[0104] As described above, the recommended information generation unit 205 may generate recommended information with content corresponding to the correlation between the frequency of the same industry and the frequency of all industries. Also, as described above, the recommended information generation unit 205 may generate recommended information with content corresponding to the correlation between the frequency in the related information and the frequency in the FB information. This makes it possible to provide the target user with a basis for judging whether to obtain the recommended registration data.

[0105] (Example of generating recommended information 2) Regarding the frequency by industry type, in each of the above examples, the candidate data evaluation unit 204 calculates the frequency of the same industry and the frequency of other industries, but the frequency may be calculated for individual industries. And in this case, the recommended information generation unit 205 may generate recommended information based on the frequency of each industry. This will be described based on FIG. 9. FIG. 9 is a diagram showing an example of generating recommended information based on the frequency of each industry.

[0106] 9, the candidate data evaluation unit 204 calculates the frequency of the entire related user information, which combines interest information and Facebook information, the frequency of interest information, and the frequency of Facebook information, for all industries. In addition, the candidate data evaluation unit 204 calculates the overall frequency, the frequency of interest information, and the frequency of Facebook information for each individual industry.

[0107] The upper table in Figure 9 shows the frequency (a%) in interest information, the frequency (b%) in Facebook information, and the frequency (c%) in interest information and Facebook information as a whole, calculated for the same industry. Also shown for industry J (an industry different from that of the target user), are the frequency (a'%) in interest information, the frequency (b'%) in Facebook information, and the frequency (c'%) in interest information and Facebook information as a whole. Furthermore, for industry K (an industry different from the target user and different from industry J), are the frequency (a''%) in interest information, the frequency (b''%) in Facebook information, and the frequency (c''%) in interest information and Facebook information as a whole.

[0108] The recommended information generator 205 generates recommended information based on these frequency values ​​calculated for each candidate data. For example, the recommended information generator 205 may compare the frequency (a%) of information of interest in the same industry with the frequency (a'%) of information of interest in industry J, as shown by the frame line Y1 in Fig. 9. If the comparison result is a'>a, the recommended information generator 205 may generate recommended information indicating that data XXXX is of higher interest in industry J than in the same industry, as shown in Fig. 9.

[0109] In addition, the recommendation information generation unit 205 may extract candidate data that is frequently or rarely occurring in some industries and generate recommendation information for the candidate data. For example, the recommendation information generation unit 205 may extract those with a frequency less than the threshold from the calculated frequencies, and generate recommendation information indicating that there is low interest or few feedbacks for the data corresponding to the extracted frequencies. For example, in the example of FIG. 9, when the frequency (a’’%) of the interest information of the data XXXX of the industry K indicated by the frame line Y2 is less than the threshold, the recommendation information generation unit 205 may generate recommendation information indicating that there is low interest in the data XXXX in the industry K.

[0110] In addition, the recommendation information generation unit 205 may generate recommendation information according to the magnitude relationship between the frequency of interest information and the frequency of FB information. For example, as shown by the frame lines Y1 and Y3 in FIG. 9, the recommendation information generation unit 205 may compare the frequency (a%) of the interest information of the same industry with the frequency (a’%) of the interest information of the industry J, and then may also compare the frequency (a’%) of the interest information of the industry J with the frequency (b’%) of the FB information of the industry J. And when the former comparison result is a’>a and the latter comparison result is a’>b’, the recommendation information generation unit 205 may generate recommendation information indicating that for the data XXXX, the interest is higher in the industry J than in the same industry, and the high interest is prominent in the industry J.

[0111] In addition, for example, as shown by the frame lines Y2 and Y4 in FIG. 9, when the frequency (a’’%) of the interest information of the industry K is below the threshold, the recommendation information generation unit 205 may compare the frequency (a’’%) of the interest information with the frequency (b’’%) of the FB information for the industry K. And when this comparison result is a’’<b’’, the recommendation information generation unit 205 may generate recommendation information indicating that for the data XXXX, although the interest in the data XXXX in the industry K is low, the amount of feedback is more prominent than the degree of interest.

[0112] (Example 3 of generating recommendation information) The information processing device 2 may allow the target user to specify conditions for generating recommended information. The specification by the target user may be received via the communication unit 22 or the input unit 23. For example, when it is specified that recommended information should be generated for related users in the same industry, the recommended information generation unit 205 generates recommended information based on related user association information of related users in the same industry in accordance with this specification. Also, for example, when an industry is specified, the recommended information generation unit 205 generates recommended information based on related user association information of related users in the specified industry. Furthermore, the target user may also specify a threshold value for determining whether or not to generate recommended information.

[0113] The generation of recommended information based on such a target user's specification will be described with reference to Fig. 10. Fig. 10 is a diagram showing an example of generation of recommended information based on a user's specification. When it is specified to generate recommended information based on related user association information of related users in the same industry, the recommended information generation unit 205 generates recommended information based on the frequency of the same industry, as shown by the frame line Z1 in Fig. 10.

[0114] In this case, if the threshold for determining whether or not to generate recommended information is designated as Th1, the recommended information generation unit 205 generates recommended information for candidate data if there is candidate data with a frequency exceeding the threshold Th1. For example, when determining whether or not to generate recommended information based on the overall frequency (c%) of interest information and FB information in the same industry, if the condition c>Th1 is satisfied for data XXXX, the recommended information generation unit 205 may generate recommended information indicating that this data is attracting interest or a lot of feedback in the same industry.

[0115] Similarly, when an industry is specified, the recommended information generator 205 generates recommended information based on the frequency of the specified industry, as shown by the frame line Z2 in FIG. 10. In this case, if the threshold for determining whether to generate recommended information is specified as Th2, the recommended information generator 205 generates recommended information for candidate data for the specified industry whose frequency exceeds the threshold Th2. For example, when determining whether to generate recommended information based on the overall frequency (c'%) of interest information and FB information, if the condition c'>Th2 is satisfied for data XXXX, the recommended information generator 205 may generate recommended information indicating that this data is attracting interest or a lot of feedback in the specified industry. Note that the thresholds Th1 and Th2 may be specified by the target user or may be predetermined.

[0116] As described above, the recommended information generation unit 205 may generate recommended information based on the related user association information of related users in a predetermined business type among the related users extracted by the related user extraction unit 202. This configuration provides the effect of being able to generate recommended information characteristic of a predetermined business type in addition to the effect achieved by the information processing device 1 according to exemplary embodiment 1. Note that the predetermined business type may be the same business type as that of the target user or may be a different business type, and the target user may be able to specify this business type as described above.

[0117] [Example of generating recommended information using preference information] The information processing device 2 can also generate recommended information according to preference information that indicates the preferences of the target user. This will be described with reference to Fig. 11. Fig. 11 is a flow diagram showing the flow of a method for generating recommended information using preference information. Note that S31 and S36 to S37 in Fig. 11 are similar to S21 and S24 to S25 in Fig. 6, respectively, and therefore the processing of S32 to S35 will be mainly described here.

[0118] In S32, the candidate data extraction unit 203 acquires preference information. The preference information is information indicating the preference of the target user, i.e., the type of data the target user prefers. For example, the preference information may be information indicating that the target user prefers data that is of high interest to other companies, or information indicating that the target user prefers data that is highly rated by other companies. Furthermore, for example, the preference information may indicate that the target user is future-oriented or innovative-oriented, or that the target user is conservative-oriented, or may indicate that the target user has a balanced preference between future-oriented or innovative-oriented and conservative-oriented.

[0119] The method for acquiring the preference information is not particularly limited. For example, the preference information may be input by the target user. Alternatively, the candidate data extraction unit 203 may determine the preference of the target user based on the target user's past behavior in the data distribution service and generate preference information based on the determination result. For example, preference information indicating a future-oriented user may be generated for a target user who selected options such as future prospects or trend watch in the data distribution service, or for a target user whose acquired data contains a high proportion of data that does not contain other companies' Facebook information. Alternatively, preference information indicating a conservative-oriented user may be generated for a target user who selected options such as warranty or safety and security, or for a target user whose acquired data contains a high proportion of data that contains other companies' Facebook information. Note that preference information does not necessarily need to be input or generated in S32. For example, preference information input or generated in advance may be stored in the storage unit 21. In this case, the preference information stored in the storage unit 21 may be acquired in S32.

[0120] In S33, the candidate data extraction unit 203 determines related user association information to be used for extracting candidate data based on the preference information acquired in S32. The related user association information corresponding to the preference information may be determined in advance. For example, when preference information indicating a future orientation is acquired, interest information may be used; when preference information indicating a conservative orientation is acquired, FB information may be used; and when preference information indicating a balanced future orientation is acquired, both interest information and FB information may be used.

[0121] In S34, the candidate data extraction unit 203 extracts candidate data based on the related user associated information determined in S33. For example, if the candidate data extraction unit 203 determines in S33 to extract candidate data using related information, it extracts data of interest indicated in the related information as candidate data. Similarly, if the candidate data extraction unit 203 determines in S33 to extract candidate data using FB information, it extracts acquired data indicated in the FB information as candidate data. Furthermore, if the candidate data extraction unit 203 determines in S33 to extract candidate data using both interest information and FB information, it extracts both the data of interest and acquired data as candidate data.

[0122] In S35, the candidate data evaluation unit 204 calculates an evaluation value for each candidate data extracted in S34. Specifically, if acquired data is extracted as candidate data in S34, the candidate data evaluation unit 204 calculates an evaluation value for the acquired data. Also, if data of interest is extracted as candidate data in S34, the candidate data evaluation unit 204 calculates an evaluation value for the data of interest. And, if acquired data and data of interest are extracted as candidate data in S34, the candidate data evaluation unit 204 calculates evaluation values ​​for the acquired data and the data of interest, respectively. The calculation of the evaluation value may be performed for each industry type and for each industry type, as in the example of FIG. 6.

[0123] In this way, by extracting candidate data based on the related user correspondence information corresponding to the preference information acquired in S32, recommended information corresponding to the preference information is generated in S37. For example, recommended information recommending data of interest to a future-oriented target user is generated, and recommended information recommending already acquired data is generated to a conservative-oriented target user.

[0124] Note that both the data of interest and the acquired data may be extracted in S34, and the candidate data evaluation unit 204 may weight the evaluation values ​​when calculating the evaluation values ​​in S35. In this case, for a future-oriented target user, the candidate data evaluation unit 204 may weight the evaluation value based on the data of interest more heavily than the evaluation value based on the acquired data. Furthermore, for a conservative-oriented target user, the candidate data evaluation unit 204 may weight the evaluation value based on the acquired data more heavily than the evaluation value based on the data of interest. Furthermore, for a balanced-oriented target user, the candidate data evaluation unit 204 may weight the evaluation value based on the acquired data to be the same or approximately the same as the evaluation value based on the data of interest. This allows the preference information to be reflected even when recommendation information is generated based on both the data of interest and the acquired data.

[0125] Here, it is unclear whether the data of interest indicated by the interest information is data that the related user has already acquired, and there is some uncertainty as to whether the data is useful to the related user. Therefore, there is some uncertainty as to whether the data of interest is useful to the target user. However, if recommended information indicating data of interest that the related user has not yet acquired is generated and presented to the target user, the target user may be able to discover the usefulness of the data before the related user. In other words, recommended information indicating data of interest can be said to be information suitable for a target user who is innovative.

[0126] On the other hand, FB information indicates the feedback content for acquired data. Therefore, by referring to the feedback content, it is easy to determine whether the acquired data is useful to the target user. Therefore, when generating recommended information indicating this acquired data, it is possible to recommend data that is likely to be useful to the target user. In other words, recommended information indicating acquired data for which feedback has been provided can be said to be information suitable for target users who are conservative.

[0127] Therefore, as described above, in the information processing device 2 according to the present exemplary embodiment, when the related user associated information includes interest information and Facebook information, the recommended information generation unit 205 may generate recommended information based on either or both of the interest information and Facebook information according to the preference information indicating the preference of the target user. This provides the effect of being able to generate recommended information suited to the preference of the target user in addition to the effect provided by the information processing device 1 according to the exemplary embodiment 1.

[0128] [Modification] The execution entity of each process described in each of the above embodiments is arbitrary and is not limited to the above examples. In other words, the devices constituting the recommendation information generation system can be changed as appropriate as long as they can execute each process described in each of the above embodiments.

[0129] 3, one information processing device 2 provides the data distribution service (such as accepting registered data and controlling transmission), extracts related users, and generates recommended information, but these functions may be performed by separate devices. For example, the recommended information generation system may include a service providing device that provides the data distribution service, a related user extraction device that extracts related users for a target user of the data distribution service based on target user association information associated with the target user, and a recommended information generation device that generates recommended information indicating data that is recommended for the target user to acquire based on the related user association information associated with the related users.

[0130] [Software implementation example] Some or all of the functions of the information processing devices 1 and 2 may be realized by hardware such as an integrated circuit (IC chip), or by software.

[0131] In the latter case, the information processing devices 1 and 2 are realized, for example, by a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in FIG. 12. The computer C includes at least one processor C1 and at least one memory C2. The memory C2 stores a program P for operating the computer C as the information processing devices 1 and 2. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing each function of the information processing devices 1 and 2.

[0132] The processor C1 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0133] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may also include a communication interface for transmitting and receiving data to and from other devices. The computer C may also include an input / output interface for connecting input / output devices such as a keyboard, mouse, display, and printer.

[0134] Furthermore, the program P can be recorded on a non-transitory tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can acquire the program P via such a recording medium M. The program P can also be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also acquire the program P via such a transmission medium.

[0135] [Appendix 1] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technical means disclosed in the above-described embodiments are also included in the technical scope of the present invention.

[0136] [Appendix 2] Some or all of the above-described embodiments can also be described as follows: However, the present invention is not limited to the following described aspects.

[0137] (Appendix 1) An information processing device comprising: a related user extraction means for extracting related users related to a target user from among a plurality of users of a data distribution service that enables other users to acquire data registered by a certain user, based on target user correspondence information associated with the target user, from the plurality of users; and a recommendation information generation means for generating recommendation information indicating data that is recommended for the target user to acquire from among data that can be acquired through the data distribution service, based on the related user correspondence information associated with the related user. With this configuration, the convenience of the data distribution service can be improved.

[0138] (Appendix 2) The information processing device according to Supplementary Note 1, wherein the target user corresponding information includes information indicating a category of data registered in the data distribution service by the target user, and the related user extraction means extracts, as the related users, users who have registered in the data distribution service data of the same or corresponding category as the data registered in the data distribution service by the target user. This configuration can increase the possibility of generating recommended information indicating registered data that is useful to the target user.

[0139] (Appendix 3) The information processing device according to claim 1, wherein the target user information includes information indicating the business type of the target user, and the related user extraction means extracts users in the same or corresponding business type as the target user as the related users. This configuration can increase the possibility of generating recommended information indicating registered data that is useful to the target user.

[0140] (Appendix 4) The information processing device according to any one of Supplementary Notes 1 to 3, wherein the related user corresponding information includes interest information indicating data of interest that is data in which the related user is interested, and the recommended information generating means generates the recommended information recommending acquisition of the data of interest. This configuration can increase the possibility of generating recommended information indicating registered data that is useful to the target user.

[0141] (Appendix 5) The information processing device according to any one of Supplementary Notes 1 to 3, wherein the related user associated information includes information indicating acquired data that the related user has acquired through the data distribution service, and the recommended information generation means generates the recommended information recommending acquisition of the acquired data. This configuration can increase the possibility of generating recommended information indicating registered data that is useful to the target user.

[0142] (Appendix 6) The information processing device according to Supplementary Note 5, wherein the related user correspondence information includes feedback information indicating feedback content from the related users regarding acquired data that the related users have acquired through the data distribution service, and the recommendation information generation means generates the recommendation information recommending acquisition of the acquired data for which feedback of predetermined content has been provided. This configuration can further increase the possibility of generating recommendation information indicating registered data that is useful to the target user.

[0143] (Appendix 7) The information processing device according to any one of Supplementary Notes 1 to 6, wherein the recommended information generation means generates the recommended information based on the related user association information of the related users in a predetermined business type among the related users extracted by the related user extraction means. This configuration makes it possible to generate recommended information that is characteristic of the predetermined business type.

[0144] (Appendix 8) The information processing device according to Supplementary Note 1, wherein the related user associated information includes interest information indicating data of interest, which is data in which the related user is interested, and feedback information indicating feedback from the related user regarding acquired data that the related user has acquired through the data distribution service, and the recommended information generation means generates the recommended information based on either or both of the interest information and the feedback information in accordance with preference information indicating the preference of the target user. This configuration makes it possible to generate recommended information that matches the preference of the target user.

[0145] (Appendix 9) A recommended information generation method including: extracting, from among a plurality of users of a data distribution service that enables other users to acquire data registered by a given user, related users associated with a target user based on target user association information associated with the target user, and generating, based on the related user association information associated with the related users, recommended information indicating data that is recommended for the target user to acquire from among data available through the data distribution service. This recommended information generation method can improve the convenience of the data distribution service.

[0146] (Appendix 10) A recommendation information generation program for causing a computer to operate as the information processing device according to Supplementary Note 1, the recommendation information generation program causing the computer to function as each of the means. This program can improve the convenience of data distribution services.

[0147] (Appendix 11) A recommendation information generation system including: a service providing device that provides a data distribution service that enables other users to acquire data registered by a certain user; a related user extraction device that extracts related users related to a target user from among multiple users of the data distribution service, based on target user association information associated with the target user; and a recommendation information generation device that generates recommendation information indicating data that is recommended for the target user to acquire from among data that can be acquired through the data distribution service, based on the related user association information associated with the related users. This recommendation information generation system can improve the convenience of the recommendation information generation system.

[0148] [Appendix 3] Some or all of the above-described embodiments can also be expressed as follows: An information processing device including at least one processor that executes, for a target user among a plurality of users of a data distribution service that allows other users to acquire data registered by a given user, a process of extracting related users related to the target user from among the plurality of users based on target user association information associated with the target user, and a process of generating recommendation information indicating data that is recommended for the target user to acquire from among data that can be acquired through the data distribution service, based on the related user association information associated with the related users.

[0149] The information processing device may further include a memory that stores a program for causing the processor to execute the process of identifying the related users and the process of generating the recommendation information. The program may also be recorded on a computer-readable, non-transitory, tangible recording medium. [Explanation of symbols]

[0150] 1. Information processing equipment 11 Related user extraction section 12 Recommendation information generation unit 100 Recommendation Generation System 2. Information processing equipment 202 Related user extraction unit 205 Recommendation information generation unit 212 Recommendations

Claims

1. a related user extraction means for extracting related users related to a target user from among a plurality of users of a data distribution service that allows other users to acquire data registered by a certain user, based on target user correspondence information associated with the target user; a recommendation information generating means for generating recommendation information indicating data that is recommended for acquisition by the target user from among data that can be acquired through the data distribution service, based on related user association information associated with the related user; The related user corresponding information includes interest information indicating data of interest, which is data that the related user has not yet acquired but is interested in; the recommendation information generating means generates the recommendation information recommending acquisition of the data of interest; The related user extraction means extracts the related users by any one of the following (1) to (5): (1) Using the target user correspondence information including information indicating the category of data registered by the target user in the data distribution service, extracting, as the related users, users who have registered in the data distribution service data of the same or corresponding category as the data registered by the target user in the data distribution service; (2) extracting, as the related users, users in the same or corresponding industries as the target users using the target user correspondence information including information indicating the industry of the target users; (3) Classifying the data registered in the data distribution service using a machine-learned classification model, and extracting, as related users, users who have registered in the data distribution service data that is classified in the same category as the data registered in the data distribution service by the target user. (4) extracting, as related users, users who have registered data in the data distribution service that contains the same words or words with the same meaning as the data registered in the data distribution service by the target user; (5) An information processing device that calculates the similarity between data registered by the target user in the data distribution service and data registered by other users in the data distribution service based on the number of the same words or words with the same meaning contained in the data, and extracts related users based on the calculated similarity.

2. the related user correspondence information includes information indicating acquired data that the related user has acquired through the data distribution service; The information processing apparatus according to claim 1 , wherein the recommendation information generating means generates the recommendation information that recommends acquisition of the acquired data.

3. the related user response information includes feedback information indicating feedback content of the related user regarding acquired data that the related user has acquired through the data distribution service; The information processing apparatus according to claim 2 , wherein the recommendation information generating means generates the recommendation information that recommends acquisition of the acquired data for which feedback of predetermined content has been provided.

4. 4. The information processing device according to claim 1, wherein the recommendation information generation means generates the recommendation information based on the related user correspondence information of the related users in a predetermined industry among the related users extracted by the related user extraction means.

5. The related user correspondence information includes interest information indicating data of interest, which is data in which the related user is interested, and feedback information indicating feedback content of the related user with respect to acquired data that the related user has acquired through the data distribution service, The information processing apparatus according to claim 1 , wherein the recommended information generating means generates the recommended information based on either or both of the interest information and the feedback information in accordance with preference information indicating the preference of the target user.

6. At least one processor extracting related users related to a target user from among a plurality of users of a data distribution service that allows other users to acquire data registered by a certain user, based on target user correspondence information associated with the target user; generating recommendation information indicating data that is recommended for acquisition by the target user from among data that can be acquired through the data distribution service, based on related user association information associated with the related user; The related user corresponding information includes interest information indicating data of interest, which is data that the related user has not yet acquired but is interested in; the at least one processor generates the recommendation information recommending acquisition of the data of interest; The at least one processor extracts the related users by any one of the following methods (1) to (5): (1) Using the target user correspondence information including information indicating the category of data registered by the target user in the data distribution service, extracting, as the related users, users who have registered in the data distribution service data of the same or corresponding category as the data registered by the target user in the data distribution service; (2) extracting, as the related users, users in the same or corresponding industries as the target users using the target user correspondence information including information indicating the industry of the target users; (3) Classifying the data registered in the data distribution service using a machine-learned classification model, and extracting, as related users, users who have registered in the data distribution service data that is classified in the same category as the data registered in the data distribution service by the target user. (4) extracting, as related users, users who have registered data in the data distribution service that contains the same words or words with the same meaning as the data registered in the data distribution service by the target user; (5) A method for generating recommended information, which calculates the similarity between data registered by the target user in the data distribution service and data registered by other users in the data distribution service based on the number of the same words or words with the same meaning contained in those data, and extracts related users based on the calculated similarity.

7. Computer, a related user extraction means for extracting related users related to a target user from among a plurality of users of a data distribution service that allows other users to acquire data registered by a certain user, based on target user correspondence information associated with the target user; a recommendation information generating unit that generates recommendation information indicating data that is recommended for the target user to acquire from among data that can be acquired through the data distribution service, based on related user association information associated with the related user; The related user corresponding information includes interest information indicating data of interest, which is data that the related user has not yet acquired but is interested in; the recommendation information generating means generates the recommendation information recommending acquisition of the data of interest; The related user extraction means extracts the related users by any one of the following (1) to (5): (1) Using the target user correspondence information including information indicating the category of data registered by the target user in the data distribution service, extracting, as the related users, users who have registered in the data distribution service data of the same or corresponding category as the data registered by the target user in the data distribution service; (2) extracting, as the related users, users in the same or corresponding industries as the target users using the target user correspondence information including information indicating the industry of the target users; (3) Classifying the data registered in the data distribution service using a machine-learned classification model, and extracting, as related users, users who have registered in the data distribution service data that is classified in the same category as the data registered in the data distribution service by the target user. (4) extracting, as related users, users who have registered data in the data distribution service that contains the same words or words with the same meaning as the data registered in the data distribution service by the target user; (5) A recommendation information generation program that calculates the similarity between data registered by the target user in the data distribution service and data registered by other users in the data distribution service based on the number of the same words or words with the same meaning contained in those data, and extracts related users based on the calculated similarity.

Images