Recommendation server, recommendation method and program

Abstract

To provide a mechanism for recommending content based on a predetermined factor which is associated with the content and a user.SOLUTION: A recommendation server includes: a first model acquisition unit which acquires a first trained estimation model which is generated by training a first estimation model using factor information associated with content and factor score information associated with a user, as explanatory variables, and reaction probability, as an objective variable; a first estimation unit which estimates, for each of multiple pieces of target content, reaction probability by inputting factor information associated with the target content and factor score information associated with a target user to the first trained estimation model; and a first extraction unit which extracts one or more pieces of target content from among the multiple pieces of target content on the basis of the reaction probability estimated for each of the multiple pieces of target content.SELECTED DRAWING: Figure 19

Description

[Technical field]

[0001] The present invention relates to a recommendation server, a recommendation method, and a program. [Background technology]

[0002] As a document disclosing the background art of this technical field, there is JP 2011-175362 A (Patent Document 1). This publication states that "recommendation engine: a system module that recommends content based on a user's preferences or a user's actions regarding the content" (see paragraph 0028).

[0003] More specifically, this publication states that "First, the analysis unit 120 acquires basic attributes of the content stored in the attribute table of the storage unit 110 (step S102). Next, the analysis unit 120 calculates attribute values ​​of the extended attributes based on the attribute values ​​of the basic attributes according to a probabilistic classification method using PLSA or LDA (step S104). The analysis unit 120 stores the attribute values ​​of the extended attributes calculated here in the attribute table" (see paragraph 0102). This publication also states that "First, the recommendation unit 140 selects content to be recommended to the user using the attribute values ​​stored in the attribute table held by the storage unit 110 (step S202)" (see paragraph 0106). That is, this publication states that attribute values ​​are calculated by PLSA and stored in the attribute table, and content to be recommended to the user is selected using the attribute values ​​stored in the attribute table. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2011-175362 A Summary of the Invention [Problem to be solved by the invention]

[0005] The PLSA described in Patent Document 1 is known as one of the recommendation algorithms. This PLSA is an algorithm that estimates the latent factors behind the user and the content after the fact (implicitly). In contrast to recommendation methods that use such algorithms, the present invention provides a mechanism for associating the user and the content with a predetermined factor and recommending the content based on the factor. [Means for solving the problem]

[0006] In order to solve the above problems, for example, the configurations described in the claims are adopted. The present application includes multiple means for solving the above-mentioned problems. One example is a recommendation server including: a first model acquisition unit that acquires a first trained estimation model generated by training a first estimation model using factor information associated with a content and factor score information associated with a user as explanatory variables and a response probability as a target variable; a first estimation unit that inputs factor information associated with a target content and factor score information associated with a target user to the first trained estimation model for each of a plurality of target contents to estimate a response probability; and a first extraction unit that extracts one or more target contents from the plurality of target contents based on the response probability estimated for each of the plurality of target contents. Effect of the Invention

[0007] According to the present invention, it is possible to provide a mechanism for associating a user with a content with a predetermined factor and recommending content based on the factor. Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0008] [Figure 1] Figure 1 shows an example of a recommendation method using Matrix Factorization. [Diagram 2] Figure 2 shows an example of a conceptual model of the overall domain and individual domains. [Diagram 3] Figure 3 shows another example of a conceptual model of the global domain and individual domains. [Figure 4] FIG. 4 shows examples of WB factors. [Diagram 5] FIG. 5 shows an example of a factor score table. [Figure 6] FIG. 6 shows a comparative example of a recommendation method. [Figure 7] FIG. 7 shows an example of a topic model. [Figure 8] FIG. 8 shows an example of the configuration of an information processing system 800. [Figure 9] FIG. 9 shows an example of the configuration of the recommendation server 801. [Figure 10] FIG. 10 shows an example of user attribute information 1000. [Figure 11] FIG. 11 shows an example of factor score information 1100. [Figure 12] FIG. 12 shows an example of a content action log 1200. [Figure 13] FIG. 13 shows an example of content behavior score information 1300. [Figure 14] FIG. 14 illustrates an example of a service action log 1400. [Figure 15] FIG. 15 shows an example of the service action score information 1500. [Figure 16] FIG. 16 shows an example of content information 1600. [Figure 17] FIG. 17 shows an example of service information 1700. [Figure 18] FIG. 18 shows an example of an activity log 1800. [Figure 19] FIG. 19 shows an example of the flow of a recommendation process. [Figure 20] FIG. 20 shows an example of this scoring process. [Figure 21] FIG. 21 shows an example of this scoring process. [Figure 22]FIG. 22 shows an example of input data to be trained by the global optimization model. [Diagram 23] FIG. 23 shows an example of input data to be input to the trained global optimization model. [Figure 24] FIG. 24 shows an example of input data to be trained by the individual optimization model. [Diagram 25] FIG. 25 shows an example of input data to be input to the trained individual optimization model. [Figure 26] FIG. 26 shows an example of a registration screen. [Figure 27] FIG. 27 shows an example of a registration screen. [Figure 28] FIG. 28 shows an example of a registration screen. [Figure 29] FIG. 29 shows an example of a recommendation screen. [Diagram 30] FIG. 30 shows an example of a recommendation screen. [Diagram 31] FIG. 31 shows an example of the details screen. [Diagram 32] FIG. 32 shows an example of My Page. [Diagram 33] FIG. 33 shows an example of a details screen. [Diagram 34] FIG. 34 shows an example of an editing screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] 1. Example Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1-1. Overview Conventionally, many recommendation services make recommendations within individual domains. For example, Figure 1 shows an example of a recommendation method using Matrix Factorization.

[0010] In the recommendation method shown in the figure, a rating matrix 101 is decomposed into a user feature matrix 102 and an item feature matrix 103. Then, the product of the two decomposed matrices is taken to generate a new rating matrix 104. Finally, items with high values ​​in the generated new rating matrix 104 are recommended. The upper part of the figure shows a method for recommending products based on product purchase data, and the lower part shows a method for recommending movies based on movie viewing data.

[0011] In such recommendation methods, user preferences are defined from behavioral data of individual domains, making it difficult to make recommendations across domains.

[0012] In contrast, the recommendation method according to the present embodiment utilizes factors expressing the user's latent psychological state, specifically, well-being factors expressing the values ​​of human happiness. Specifically, in this recommendation method, well-being factors defined in advance by the business operator are associated with each of the user, well-being content, and well-being service. Then, from the perspective of maximizing the user's well-being, optimization of the entire domain is performed using well-being content, and then optimization of individual domains is performed using well-being services. Therefore, this recommendation method makes it possible to make recommendations across domains.

[0013] The above-mentioned well-being is a term that means health and security, a satisfying living condition, happiness, etc. In the following explanation, this well-being may be abbreviated as "WB."

[0014] 2 shows an example of a conceptual model of an overall domain and an individual domain. In the conceptual model shown in the figure, WB contents 202 (e.g., “Work”) belonging to an overall domain 201 are associated with one or more WB services 204 (e.g., “Facility Environment”) belonging to an individual domain 203.

[0015] The recommendation method according to the present embodiment will be explained using this conceptual model. First, as a global optimization, "work" is recommended to a user who has a high interest in "work." Then, as an individual optimization, a specific service related to "work" is recommended to the user.

[0016] 3 shows another example of a conceptual model of the overall domain and individual domains. In the conceptual model shown in the figure, WB contents 302 belonging to an overall domain 301 are associated with one or more WB services 304 belonging to individual domains 303.

[0017] The recommendation method according to the present embodiment will be explained using this conceptual model. First, as a global optimization, WB content is recommended to the user. At this time, the consciousness area in which the user is highly interested is estimated. Next, as an individual optimization, a WB service associated with the WB content is recommended to the user. In other words, a WB service is further recommended within the user's area of ​​interest. This improves the well-being of the user.

[0018] Next, the WB factor will be explained. Figure 4 shows examples of WB factors. The total of 21 WB factors shown in the figure are 7 WB factors related to activity or body that belong to the category "wellness", 7 WB factors related to element or mind that belong to the category "newness", and 7 WB factors related to state or society that belong to the category "community".

[0019] These WB factors are defined in advance by the business operator as latent factors from the viewpoint of well-being. In other words, these WB factors are defined in advance by the business operator as expressing the user's happiness level or expressing the user's values ​​regarding happiness level.

[0020] These WB factors are associated with a user. As a specific method, first, a questionnaire about WB factors is given to the user when registering for the service. This questionnaire asks the user to indicate the importance of each WB factor. For details of this questionnaire, see, for example, a registration screen 2800 shown in FIG. 28.

[0021] The results of the questionnaire are then stored in a table in association with the user. An example of the factor score table is shown in Fig. 5. The factor score table 500 shown in the figure stores the scores of each WB factor for each user.

[0022] Next, the WB content will be described. The WB content is recommendation content defined in advance by the business operator, and is content for clarifying the user's current interests and areas of interest. In other words, the WB content is advice information defined in advance to improve the user's happiness. As an example, 400 types of this WB content are prepared. For specific examples of this WB content, see, for example, the recommendation screen 2900 shown in FIG. 29.

[0023] Each WB content is associated with a WB factor in a table. For a table associating WB contents with WB factors, see, for example, content information 1600 shown in FIG.

[0024] Next, the WB service will be described. The WB service is a provider service for recommendation that is predefined by the provider, and is a service within the area of ​​interest and concern of each user at that time. For a specific example of this WB service, see, for example, the recommendation screen 3000 shown in FIG.

[0025] Each WB service is associated with a WB factor in a table. For a table associating WB services with WB factors, see, for example, service information 1700 shown in FIG. Furthermore, each WB service is associated with a WB content, as shown in FIG. 2 and FIG.

[0026] Next, the flow of the recommendation method according to this embodiment will be described. In the recommendation method according to the present embodiment, first, the user's interests and areas of interest are estimated in the entire domain, and WB content is recommended. Next, WB services within the user's interests and areas of interest are recommended in individual domains. In other words, WB services associated with the WB content to which the user responded are recommended in the individual domains.

[0027] As described above, the recommendation method according to the present embodiment has two stages. In contrast, a recommendation method that directly recommends a WB service without recommending a WB content can be considered. Figure 6 shows a comparison example between the former recommendation method and the latter recommendation method.

[0028] All of the recommendation methods shown in the figure are methods for recommending books. However, stepwise recommendation 601 first recommends a book genre, and then recommends books that belong to that genre. In contrast, non-stepwise recommendation 602 directly recommends a book without recommending a book genre.

[0029] The two-step recommendation method such as stepwise recommendation 601 is based on the empirical rule that users respond better to gradually narrowing down their interests and areas of concern to more specific WB services than to suddenly recommending WB services that the user may not be interested in.

[0030] Next, the difference between the recommendation method and the topic model according to the present embodiment will be described. First, a topic model is one of the recommendation algorithms, and is an algorithm that retroactively (implicitly) estimates the latent factors behind users and recommended content. A representative example of this topic model, PLSA (Probabilistic Latent Semantic Analysis), is illustrated in Figure 7 and expressed by the following formula.

number

[0031] In this formula (1), P(c i ,g j ) represents the co-occurrence probability, and P(z k ), P(c i |z k ) and P(g j |z k ) represents a random variable. In addition, in this formula (1) and Figure 7, c i represents user information, and g j represents content information, and z k represents a latent factor. This PLSA uses the EM algorithm to calculate the co-occurrence probability P(c i ,g j ) the latent factor z that maximizes k is calculated a posteriori (implicitly).

[0032] In contrast, in the recommendation method according to the present embodiment, the latent factor z k The information equivalent to the above is defined in advance by the business operator and is associated with users and content. In this respect, this recommendation method differs from PLSA. In addition, the latent factor z k The method of associating information corresponding to the latent factor z with a user may be, for example, by asking the user (specifically, by conducting a questionnaire). k The method of associating information equivalent to the above with the content may be based on empirical rules, for example.

[0033] 1-2.Configuration Next, an information processing system according to an embodiment will be described. Fig. 8 shows an example of the configuration of an information processing system. The information processing system 800 shown in the figure includes a recommendation server 801 and multiple user terminals 802. Each component device is connected via a wired or wireless network and can transmit and receive information to and from each other.

[0034] A recommendation server 801 included in this information processing system 800 is a server for recommending WB contents and WB services to users. On the other hand, the user terminal 802 is a terminal device for receiving the recommendation service provided by the recommendation server 801 .

[0035] Each of the components of the information processing system 800 may be, for example, a mobile terminal such as a smartphone, tablet, mobile phone, or personal digital assistant (PDA), or may be a wearable terminal such as glasses, wristwatch, or clothing. Each of the components may be a stationary or mobile computer, or a server located on the cloud or a network. Each of the components may function as a VR (Virtual Reality) terminal, an AR (Augmented Reality) terminal, or an MR (Mixed Reality) terminal. Alternatively, each of the components may be a combination of a plurality of these terminals. For example, a combination of one smartphone and one wearable terminal may function as one logical terminal. Each of the components may be an information processing terminal other than these.

[0036] Each of the constituent devices includes a processor that executes an operating system, applications, programs, etc., a main storage device such as a RAM (Random Access Memory), an auxiliary storage device such as an IC card, a hard disk drive, an SSD (Solid State Drive), a flash memory, etc., a communication control unit such as a network card, a wireless communication module, a mobile communication module, etc., an input device such as a touch panel, a keyboard, a mouse, a voice input, an input by motion detection by imaging of a camera unit, etc., and an output device such as a monitor, a display, etc. The output device may be a device or a terminal that transmits information to be output to an external monitor, display, projector, printer, device, etc.

[0037] The main memory stores various programs and applications (modules), and the processor executes these programs and applications to realize the various functional elements of the overall system. Each of these modules may be implemented in hardware by integration or the like. Each module may be an independent program or application, or may be implemented as a subprogram or function within a single integrated program or application.

[0038] In this specification, each module is described as an entity (subject) that performs processing, but in reality, a processor that processes various programs, applications, etc. (modules) executes the processing.

[0039] Various databases (DB) are stored in the auxiliary storage device. A "database" is a functional element (storage unit) that stores a data set so that it can handle any data operation (e.g., extraction, addition, deletion, overwriting, etc.) from a processor or an external computer. There are no limitations on the method of implementing the database, and it may be, for example, a database management system, spreadsheet software, or a text file such as XML or JSON.

[0040] Among the components of the information processing system 800, the recommendation server 801 will be particularly described in detail below.

[0041] Fig. 9 shows an example of the configuration of a recommendation server 801. The recommendation server 801 shown in the figure is configured, for example, by one or more servers arranged on a cloud. The main memory device 901 of this server stores programs such as a model learning module 911, a scoring module 912, a reranking module 913, an extraction module 914, a screening module 915, a recommendation module 916, and an action log management module 917 as modules for overall optimization.

[0042] The main memory device 901 also stores programs such as a model learning module 921, a scoring module 922, a filtering module 923, a reranking module 924, an extraction module 925, a screening module 926, a recommendation module 927, and an action log management module 928 as modules for individual optimization.

[0043] Furthermore, the main memory device 901 stores programs such as a my page module 931 and a user information management module 932 . The above-described programs are executed by the processor 903 to realize the respective functional elements of the recommendation server 801. Each module will be described below.

[0044] The model learning module 911 learns an estimation model using WB factor information previously associated with WB content and WB factor score information previously associated with a user as explanatory variables and a reaction probability as a response variable. The module generates and acquires a learned estimation model through this learning.

[0045] The estimation model that this module trains is FM (Factorization Machines). FM has the following features: · FM is a general classifier that allows reliable parameter estimation even for very sparse data. · FM is capable of solving regression, classification and ranking problems. · FM is a model that can approximately take into account the interactions between all features. · FM has an internal mechanism to reduce the amount of calculations.

[0046] The mathematical model of this FM is generally expressed by the following equation:

number

number

[0047] Next, the scoring module 912 inputs WB factor information pre-associated with the WB content and WB factor score information pre-associated with the target user into a learned estimation model for each of the multiple WB contents, and estimates a response probability.

[0048] The reranking module 913 excludes WB content previously selected by the target user from among the WB contents whose response probabilities have been estimated. This is to avoid re-recommending WB content previously recommended. In addition, if there is a WB content similar to the WB content with a high response probability, the module rearranges the order of one or more of those WB contents. This is to avoid recommending multiple similar WB contents at once.

[0049] The extraction module 914 extracts one or more WB contents from the reranked WB contents based on the response probability estimated for each WB content. The extracted WB contents are WB contents with the top five response probabilities.

[0050] The screening module 915 filters out inappropriate content from the extracted WB content, where inappropriate content is WB content that is identified in a rule-based or model-based manner based on the user attributes of the target user.

[0051] The recommendation module 916 recommends the WB content that passes the screening to the target user.

[0052] The action log management module 917 records the target user's actions with respect to the recommended WB content.

[0053] The model learning module 921 learns an estimation model using factor information previously associated with a WB service and WB factor score information previously associated with a user as explanatory variables and a response probability as a response variable. At this time, the module learns an estimation model for each individual area. Through this learning, the module generates and acquires a learned estimation model for each individual area. The estimation model learned by this module is FM.

[0054] The scoring module 922 inputs factor information associated with the WB service in advance and WB factor score information associated with the target user in advance into a trained estimation model for each of a plurality of WB services to estimate a reaction probability. Note that the trained estimation model used in this estimation corresponds to the individual domain to which the WB service belongs.

[0055] The filtering module 923 excludes WB services that do not correspond to the WB content selected by the target user from among the WB services for which the response probability has been estimated. This is to avoid recommending WB services that are not related to the WB content selected by the target user. Note that a WB service that does not correspond to the WB content selected by the user is a WB service that does not share an individual area with the WB content selected by the user.

[0056] The reranking module 924 removes WB services that the target user previously selected from the filtered WB services. This is to avoid re-recommending WB services that have been previously recommended. In addition, if there are WB services similar to the WB services with the highest response probability, the reranking module rearranges the order of one or more of those WB services. This is to avoid recommending multiple similar WB services at once.

[0057] The extraction module 925 extracts one or more WB services from the reranked WB services based on the response probability estimated for each WB service. The extracted WB services are WB services with the top five response probabilities.

[0058] The screening module 926 filters out inappropriate WB services from the extracted WB services, where the inappropriate WB services are identified in a rule-based or model-based manner based on the user attributes of the target user.

[0059] The recommendation module 927 recommends the WB services that have passed the screening to the target user.

[0060] The action log management module 928 records the target user's actions with respect to the recommended WB service.

[0061] The My Page module 931 displays My Page (see FIG. 32) on the user terminal 802 in response to a user request.

[0062] The user information management module 932 manages user attribute information and WB factor scores.

[0063] Next, the auxiliary storage device 902 of the recommendation server 801 will be described. The auxiliary storage device 902 stores various types of information. Specifically, the device stores information such as user attribute information 1000, factor score information 1100, content behavior log 1200, content behavior score information 1300, service behavior log 1400, service behavior score information 1500, content information 1600, service information 1700, and activity log 1800.

[0064] This device also stores various data such as screen data 941 and image data 942 . Each piece of information will be explained below.

[0065] 10 shows an example of user attribute information 1000. The user attribute information 1000 shown in the figure is information in a table format showing attribute information of each user. This user attribute information 1000 has columns such as a user ID 1001, a date of birth 1002, a gender 1003, a weight 1004, a height 1005, and an interest ID 1006.

[0066] 11 shows an example of factor score information 1100. The factor score information 1100 shown in the figure is information in a table format indicating a user's WB factor scores. This factor score information 1100 has columns such as a user ID 1101, a registration date and time 1102, and a factor score 1103. Of these columns, the factor score 1103 stores a value indicating the degree to which the user attaches importance to the WB factor. As an example, the stored value is a value from "1" to "5." This factor score information 1100 is generated for each user.

[0067] FIG. 12 shows an example of a content behavior log 1200. The content behavior log 1200 shown in the figure is information in a table format showing a user's behavior history with respect to recommended WB content. This content behavior log 1200 has columns such as a user ID 1201, a content ID 1202, a display position 1203, a behavior score 1204, and a registration date and time 1205. Among these columns, the display position 1203 stores a value indicating the order in which the WB content was displayed on the recommendation screen (see FIG. 29). Furthermore, the behavior score 1204 stores a value indicating the user's reaction probability to the WB content. Specifically, the behavior score 1204 stores a value of "1" indicating that the user selected the WB content, or a value of "0" indicating that the user did not select the WB content.

[0068] At the beginning of the service, this content action log 1200 is accumulated by randomly delivering WB content to users. Then, when a certain amount of action logs are accumulated and model learning becomes possible, WB content estimated by the learned model is delivered and accumulated.

[0069] 13 shows an example of content behavior score information 1300. The content behavior score information 1300 shown in the figure is information in a table format showing the behavior score recorded or estimated for each WB content. In this information, the WB contents are sorted in descending order of behavior score, and the information shows the ranking of the WB contents.

[0070] This content behavior score information 1300 has columns such as a user ID 1301, a content ID 1302, and a behavior score 1303. Of these columns, the behavior score 1303 stores a value indicating the probability of a user's reaction to the WB content. As an example, the stored value is a value between "0" and "1." This content behavior score information 1300 is generated for each user.

[0071] FIG. 14 shows an example of a service action log 1400. The service action log 1400 shown in the figure is information in a table format showing a user's action history with respect to a recommended WB service. This service action log 1400 has columns such as a user ID 1401, a service ID 1402, a display position 1403, an action score 1404, and a registration date and time 1405. Among these columns, the display position 1403 stores a value indicating the order in which the WB service is displayed on the recommendation screen (see FIG. 30). Furthermore, the action score 1404 stores a value indicating the user's reaction probability to the WB service. Specifically, the action score 1404 stores a value of "1" indicating that the user selected the WB service, or a value of "0" indicating that the user did not select the WB service.

[0072] This service action log 1400 is accumulated by randomly delivering WB services to users at the beginning of the service. Then, when a certain amount of action logs are accumulated and model learning becomes possible, the log is accumulated by delivering WB services estimated by the learned model.

[0073] 15 shows an example of service action score information 1500. The service action score information 1500 shown in the figure is information in a table format showing the action scores recorded or estimated for each WB service. In this information, the WB services are sorted in descending order of action score, and the information shows the ranking of the WB services.

[0074] This service action score information 1500 has columns such as a user ID 1501, a service ID 1502, and an action score 1503. Of these columns, the action score 1503 stores a value indicating the probability of a user's reaction to a WB service. As an example, the stored value ranges from "0" to "1." This service action score information 1500 is generated for each user.

[0075] FIG. 16 shows an example of content information 1600. The content information 1600 shown in the figure is information in a table format showing information on each WB content. This content information 1600 has columns such as a content ID 1601, details 1602, an image ID 1603, a factor ID 1604, and an individual area ID 1605. Of these columns, the image ID 1603 stores the ID of an image of the WB content (see FIG. 29). Furthermore, the factor ID 1604 stores the ID of a WB factor previously associated with the WB content. Furthermore, the individual area ID 1605 stores the ID of an individual area previously associated with the WB content. Here, an individual area refers to a category to which the WB content or the WB service belongs. The categories include "sleep," "eat," and "play."

[0076] Fig. 17 shows an example of service information 1700. The service information 1700 shown in the figure is information in a table format showing information on each WB service. This service information 1700 has columns such as a service ID 1701, a content 1702, an image ID 1703, a factor ID 1704, and an individual area ID 1705. Of these columns, the image ID 1703 stores the ID of an image of the WB service (see Fig. 30). Furthermore, the factor ID 1704 stores the ID of a WB factor previously associated with the WB service. Furthermore, the individual area ID 1705 stores the ID of an individual area previously associated with the WB service.

[0077] Here, a method for associating a WB service with a WB factor will be described. The first method is to associate the WB factor of the WB content corresponding to the WB service with the WB service. Note that the WB content corresponding to the WB service is the WB content that shares an individual area with the WB service. The second method is to pick up the WB factors of the WB content that correspond to the WB service, and then manually assign the most important ones. This second method has an advantage over the first method in that it can reliably eliminate WB factors that are not related to the WB service.

[0078] Next, Fig. 18 shows an example of an activity log 1800. The activity log 1800 shown in Fig. 18 is information in a table format showing a user's daily activity records. This activity log 1800 has columns such as a user ID 1801, a registration date and time 1802, a WB degree 1803, and a memo 1804. Of these columns, the WB degree 1803 stores a value indicating the user's happiness or satisfaction. Furthermore, the memo 1804 stores information recorded by the user as a diary.

[0079] 1-3.Operation Next, the operation of the information processing system 800 will be described. 1-3-1. Recommendation processing FIG. 19 shows an example of the flow of a recommendation process. First, when a user wishes to receive a recommendation service, the user pre-registers with the recommendation server 801. At that time, the user information management module 932 of the recommendation server 801 causes the user terminal 802 to display a registration screen.

[0080] 26 to 28 show examples of registration screens. A registration screen 2600 shown in Fig. 26 is a screen for registering a user profile. This screen has a date of birth selection field 2601, a gender selection field 2602, a weight selection field 2603, a height selection field 2604, and a continue button 2605. When the user selects a date of birth, etc. on this screen and selects the continue button 2605, a registration screen 2700 shown in Fig. 27 is then displayed.

[0081] The registration screen 2700 is a screen for registering the user's interests. This screen has a plurality of selection buttons 2701 for selecting interests and a continue button 2702. When the user selects the selection button 2701 for his / her own interests on this screen and then selects the continue button 2702, a registration screen 2800 shown in FIG. 28 is then displayed.

[0082] The registration screen 2800 is a questionnaire screen regarding WB factors. This questionnaire screen allows the user to answer the importance of each WB factor on a five-point scale. The importance answered by the user is registered in the recommendation server 801 as a WB factor score.

[0083] This registration screen 2800 has a plurality of answer selection fields 2801 provided for each question, and a continue button 2802. When the user selects an answer to a question on this screen and selects the continue button 2802, the next questionnaire screen (not shown) is displayed.

[0084] When the user answers the questions regarding all WB factors and selects a send button (not shown), user registration is completed. In addition to the questions about WB factors, the questions presented to the user may include questions about the user's attributes. For example, the questions may include questions about the user's diet and diseases.

[0085] The information selected by the user is registered in the recommendation server 801. At that time, the user information management module 932 of the recommendation server 801 adds the profile information selected on the registration screen 2600 to the user attribute information 1000 in association with the user ID (step 1901). The profile information here refers to the date of birth, sex, weight, and height.

[0086] The module also adds the ID of the subject of interest selected on the registration screen 2700 to the user attribute information 1000 in association with the user ID (step 1901). The module also adds the WB factor scores selected on the registration screen 2800 to the factor score information 1100 in association with the user ID and the registration date and time (step 1902). This concludes the explanation of user registration.

[0087] Next, the behavior score update process will be described. The behavior score update process is a batch process that is executed periodically, such as daily, weekly, or monthly. This update process is performed for each of the WB content and the WB service. In each update process, an estimation model is learned, and a behavior score is estimated based on the learned estimation model. First, the update process performed for the WB content is described below.

[0088] The model learning module 911 of the recommendation server 801 learns an overall optimal model based on the content behavior log 1200, etc. (step 1903). At this time, the module learns the overall optimal model using the attribute information and initial WB factor score of each user and the factor information of each WB content as explanatory variables, and the behavior score regarding each WB content as a target variable. As a result of this learning, the module generates a learned overall optimal model.

[0089] 22 shows an example of input data to be trained by the global optimization model. The input data 2200 shown in the figure adopts the data format for FM. This input data 2200 has columns such as user 2201, item 2202, U_WB 2203, I_WB 2204, Y 2205, etc. Each column will be explained below.

[0090] A flag for identifying a user is stored in user 2201. For example, a record in which a flag "1" is stored in the user1 column is an action log of user 1. A flag for identifying WB content is stored in item2202. For example, a record in which a flag "1" is stored in the item1 column is an action log for WB content 1.

[0091] The user's WB factor scores are stored in U_WB 2203. Each WB factor score indicates the importance of the WB factor to the user, and is expressed in the following five levels. 1: Not at all 2: I don't think so 3: I can't decide 4: I think so 5: Strongly agree

[0092] A flag for identifying a WB factor associated with a WB content is stored in I_WB 2204. For example, if a flag "1" is stored in the item1 column for a certain record and a flag "1" is stored in the I_WB1 column, this indicates that WB content 1 is associated with WB factor 1. Note that a plurality of WB factors may be associated with one WB content.

[0093] A flag indicating an action score is stored in Y2205. The stored flags include a flag "1" indicating that a WB content has been selected, and a flag "0" indicating that a WB content has not been selected.

[0094] 22 does not include a column for user attributes. However, a column for user attributes may be included in the input data 2200. This allows the attribute information of each user to be used as an explanatory variable to be learned by the global optimization model.

[0095] Next, we will explain the estimation of behavior scores using the trained global optimal model. The scoring module 912 of the recommendation server 801 executes a scoring process using the generated trained global optimal model (step 1904). This scoring process is executed for each user.

[0096] 20 shows an example of this scoring process. In the scoring process 2000 shown in the figure, the module adds the behavior score of the WB content recommended to the user to the content behavior score information 1300 (step 2001). Here, the WB content recommended to the user refers to the WB content associated with the user in the content behavior log 1200. That is, the module identifies the WB content associated with the user in the content behavior log 1200, and adds the behavior score of the identified WB content to the content behavior score information 1300.

[0097] Next, the module estimates the behavior score of the WB content that has not been recommended to the user by using the trained global optimum model (step 2002). Here, the WB content that has not been recommended to the user refers to the WB content that is included in the content information 1600 and is not associated with the user in the content behavior log 1200. When the module identifies the WB content that has not been recommended, the module inputs the factor information of the identified WB content, the attribute information of the target user, and the initial WB factor score into the trained global optimum model to estimate the behavior score. The module adds the estimated behavior score to the content behavior score information 1300.

[0098] Fig. 23 shows an example of input data to be input to the trained global optimization model. The input data 2300 shown in the figure adopts a data format for FM. This input data 2300 has columns such as user 2301, item 2302, U_WB 2303, I_WB 2304, Y 2305, etc. The attributes of these columns are the same as those of the columns of the same name in the above-mentioned input data 2200, so their explanation will be omitted. Note that since this input data 2300 is data for estimating a behavior score, no flags are stored in Y 2305.

[0099] 23 does not include a column of user attributes. However, the input data 2300 may include a column of user attributes. This makes it possible to estimate a behavior score using attribute information of a target user as an explanatory variable.

[0100] After estimating the behavior score, the scoring module 912 sorts the WB contents in the content behavior score information 1300 in descending order of behavior score (step 2003). As a result, ranking information of the WB contents is generated in the order of the highest probability of reaction from the target user.

[0101] Next, the update process executed for the WB service will be described. The model learning module 921 of the recommendation server 801 learns an individual optimal model based on the service action log 1400, etc. (step 1905). At that time, the module learns an individual optimal model for each individual area, and generates a trained individual optimal model for each individual area. Specifically, the module learns an individual optimal model for each individual area, using attribute information and initial WB factor scores of each user and factor information of each WB content belonging to the individual area as explanatory variables, and the behavior score for each WB content as a target variable. As a result of this learning, the module generates a trained individual optimal model for each individual area.

[0102] 24 shows an example of input data to be trained by an individual optimization model. The input data 2400 shown in the figure adopts a data format for FM. This input data 2400 has columns such as user 2401, item 2402, U_WB 2403, I_WB 2404, Y 2405, etc. Each column will be explained below.

[0103] A flag for identifying a user is stored in user 2401. For example, a record in which a flag "1" is stored in the user1 column is an action log of user 1. A flag for identifying a WB service is stored in item2402. For example, a record in which a flag "1" is stored in the item1 column is an action log for WB service 1.

[0104] The user's WB factor scores are stored in U_WB 2403. Each WB factor score indicates the importance of the WB factor to the user, and is expressed in the following five levels. 1: Not at all 2: I don't think so 3: I can't decide 4: I think so 5: Strongly agree

[0105] A flag for identifying a WB factor associated with a WB service is stored in I_WB 2404. For example, if a flag "1" is stored in the item1 column for a certain record and a flag "1" is stored in the I_WB1 column, this indicates that WB service 1 is associated with WB factor 1. Note that a plurality of WB factors may be associated with one WB service.

[0106] A flag indicating an action score is stored in Y2405. The stored flags include a flag "1" indicating that the WB service has been selected, and a flag "0" indicating that the WB service has not been selected.

[0107] 24 does not include a column of user attributes. However, a column of user attributes may be included in the input data 2400. This allows the attribute information of each user to be used as an explanatory variable to be trained in an individual optimization model.

[0108] Next, the estimation of behavior scores using the trained individual optimal model will be described. The scoring module 922 of the recommendation server 801 executes a scoring process using the generated trained individual optimal model (step 1906). This scoring process is executed for each user.

[0109] 21 shows an example of this scoring process. In the scoring process 2100 shown in the figure, the module adds the behavior score of the WB service recommended to the user to the service behavior score information 1500 (step 2101). Here, the WB service recommended to the user is the WB service associated with the user in the service behavior log 1400. That is, the module identifies the WB service associated with the user in the service behavior log 1400, and adds the behavior score of the identified WB service to the service behavior score information 1500.

[0110] Next, the module estimates the behavior score of the WB service that has not been recommended to the user by using the learned individual optimal model (step 2102). Here, the WB service that has not been recommended to the user is a WB service that is included in the service information 1700 and is not associated with the user in the service action log 1400. When the module identifies a WB service that has not been recommended, the module inputs the factor information of the identified WB service, the attribute information of the target user, and the initial WB factor score into the learned individual optimal model to estimate the behavior score. Note that the learned individual optimal model used in this estimation corresponds to the individual area to which the identified WB service belongs. The module adds the estimated behavior score to the service behavior score information 1500 .

[0111] Fig. 25 shows an example of input data to be input to a trained individual optimization model. The input data 2500 shown in the figure adopts a data format for FM. This input data 2500 has columns such as user 2501, item 2502, U_WB 2503, I_WB 2504, Y 2505, etc. The attributes of these columns are the same as those of the columns of the same name in the above-mentioned input data 2400, so their explanation will be omitted. Note that since this input data 2500 is data for estimating a behavior score, no flags are stored in Y 2505.

[0112] 25 does not include a column of user attributes. However, the input data 2500 may include a column of user attributes. This makes it possible to estimate a behavior score using attribute information of a target user as an explanatory variable.

[0113] After estimating the behavior score, the scoring module 922 sorts the WB services in the service behavior score information 1500 in descending order of behavior score (step 2103). As a result, ranking information of the WB services is generated in order of the highest reaction probability of the target user. This concludes the description of the behavior score update process.

[0114] Next, a recommendation flow of WB contents will be explained. This recommendation flow starts when a user accesses a site provided by the recommendation server 801.

[0115] First, the reranking module 913 of the recommendation server 801 removes WB content previously selected by the target user from the content behavior score information 1300 of the target user (step 1907). This is to avoid re-recommending WB content previously recommended by the target user. The WB content previously selected by the target user refers to the WB content associated with the target user and a behavior score of "1" in the content behavior log 1200.

[0116] Furthermore, if there is a WB content similar to the WB content with a high behavioral score, the module rearranges the order of one or more of those WB contents (step 1907). This is to avoid recommending multiple similar WB contents at once. Note that WB contents with a high behavioral score refer, for example, to WB contents in the top five. Also, similar WB contents refer, for example, to WB contents that share a common individual area. Also, rearranging the order refers, for example, to demoting one of two similar WB contents to sixth place or lower.

[0117] Next, the extraction module 914 of the recommendation server 801 extracts the top five WB contents from the reranked content behavior score information 1300 (step 1908). Note that the number of WB contents to be extracted may be changed as appropriate.

[0118] Next, the screening module 915 of the recommendation server 801 removes inappropriate content from the extracted WB content (step 1909). Here, the inappropriate content refers to WB content that is specified on a rule-based or model-based basis based on the user attribute information 1000 of the target user.

[0119] Next, the recommendation module 916 of the recommendation server 801 recommends the WB content that has passed the screening to the target user (step 1910). At that time, the module generates a recommendation screen with reference to the content information 1600, and displays it on the user terminal 802. Fig. 29 shows an example of the recommendation screen.

[0120] The recommendation screen 2900 shown in the figure presents five WB contents 2901. The five WB contents 2901 are arranged in order of behavior score, and each is composed of a WB content description 2902 and an image 2903. By referring to this recommendation screen 2900, the user can obtain advice for improving their own well-being.

[0121] When the user selects a WB content 2901 of particular interest on the recommendation screen 2900, a recommendation flow for the WB service is started. In response to the selection, the action log management module 917 of the recommendation server 801 adds information about the selected WB content 2901 to the content action log 1200 in association with the user ID and the action score "1" (step 1911).

[0122] In addition, in response to the selection, the user information management module 932 updates the factor score information 1100 of the target user (not shown). Specifically, the module identifies the WB factor of the selected WB content 2901, and increments the score of the identified WB factor by, for example, "0.1". In this way, the user's WB factor score changes depending on the selected WB content 2901.

[0123] Returning to the explanation of the recommendation screen 2900, A user can select multiple WB contents 2901 on a recommendation screen 2900. WB contents 2901 that are not selected by the user for a certain period of time on this screen are registered in the content behavior log 1200 (step 1911). At that time, the behavior log management module 917 adds information about the unselected WB contents 2901 to the content behavior log 1200 in association with a user ID and a behavior score of "0." The above is an explanation of the recommendation flow for WB content.

[0124] Next, a recommendation flow of the WB service will be described. This recommendation flow starts when a WB content 2901 is selected on a recommendation screen 2900 illustrated in FIG.

[0125] First, the filtering module 923 of the recommendation server 801 excludes WB services that do not correspond to the WB content selected by the target user from the service action score information 1500 of the target user (step 1912). This is to avoid recommending WB services that are not related to the WB content selected by the target user. Note that a WB service that does not correspond to the WB content selected by the user is a WB service that does not share an individual area with the WB content selected by the user. Such a WB service can be identified by referring to the content information 1600 and the service information 1700.

[0126] Note that instead of excluding a WB service that does not correspond to the WB content selected by the user, the filtering module 923 may set the behavior score of the WB service to "0" to prevent the recommendation of the WB service.

[0127] Next, the reranking module 924 of the recommendation server 801 removes WB services previously selected by the user from the filtered service action score information 1500 (step 1913). This is to avoid re-recommending WB services previously recommended by the user. The WB services previously selected by the user are those associated with the user and an action score of "1" in the service action log 1400.

[0128] Furthermore, if there is a WB service similar to the WB service with a high behavior score, the module rearranges the order of one or more of those WB services (step 1913). This is to avoid recommending multiple similar WB services at once. Note that a WB service with a high behavior score is, for example, a WB service in the top five. Also, a similar WB service is, for example, a WB service that shares a common individual area. Also, rearranging the order is, for example, demoting one of two similar WB services to sixth place or lower.

[0129] Next, the extraction module 925 of the recommendation server 801 extracts the top five WB services from the reranked service action score information 1500 (step 1914). Note that the number of WB services to be extracted may be changed as appropriate.

[0130] Next, the screening module 926 of the recommendation server 801 removes inappropriate WB services from the extracted WB services (step 1915). Here, the inappropriate WB services are those that are specified on a rule-based or model-based basis based on the user attribute information 1000 of the target user.

[0131] Next, the recommendation module 927 of the recommendation server 801 recommends the WB service that passed the screening to the target user (step 1916). At that time, the module generates a recommendation screen with reference to the service information 1700, and displays it on the user terminal 802. Fig. 30 shows an example of this recommendation screen.

[0132] The recommendation screen 3000 shown in the figure presents five WB services 3001. The five WB services 3001 are arranged in order of behavior score, and each is composed of a WB service content 3002 and an image 3003. By referring to this recommendation screen 3000, the user can learn about services that can improve their own well-being.

[0133] When the user selects a WB service 3001 of particular interest on this recommendation screen 3000, the recommendation module 927 causes a details screen to be displayed on the user terminal 802. Fig. 31 shows an example of this details screen.

[0134] A details screen 3100 shown in the figure has a detailed explanation field 3101 for the WB service 3001 and a "Try it" button 3102. When the user selects the "Try it" button 3102 on this screen, the screen transitions to the site of the business providing the WB service 3001. The recommendation module 927 may directly transition from the recommendation screen 3000 to the site of the business operator without displaying the details screen 3100.

[0135] When the WB service 3001 is selected on the recommendation screen 3000, the action log management module 928 of the recommendation server 801 adds information about the selected WB service 3001 to the service action log 1400 in association with the user ID and the action score "1" (step 1917).

[0136] A user can select multiple WB services 3001 on the recommendation screen 3000. Any WB services 3001 that are not selected by the user for a certain period of time on this screen are registered in the service action log 1400 (step 1917). At that time, the action log management module 928 adds information about the unselected WB services 3001 to the service action log 1400 in association with the user ID and an action score of "0." The above is an explanation of the recommendation flow of the WB service.

[0137] According to the recommendation process described above, first, the overall domain optimization is performed using WB contents, and then the individual domain optimization is performed using WB services. This makes it possible to make recommendations across domains.

[0138] 1-3-2. My Page display process Next, the My Page display process will be described. Here, the My Page is a screen for displaying and recording the user's daily activities. The My Page module 931 of the recommendation server 801 displays the My Page on the user terminal 802 in response to a user request. At that time, the module generates the My Page based on the factor score information 1100 and the activity log 1800 of the target user.

[0139] Fig. 32 shows an example of MyPage. MyPage 3200 shown in the figure has a content confirmation button 3201, a factor score display area 3202, a calendar 3203, a MyPage button 3204, and a recommendation button 3205. Each display element will be described below.

[0140] The content confirmation button 3201 is a button for displaying a WB content recommendation screen (see FIG. 29).

[0141] The factor score display area 3202 is an area where the user's WB factor scores are displayed. In this area, category selection tabs 3206 to 3208, value selection buttons 3209 to 3211, and a radar chart 3212 are displayed. Of these display elements, the category selection tabs 3206 to 3208 are tabs for selecting the category of the WB factor to be displayed in the radar chart 3212 from among "wellness," "newness," and "community." The value selection buttons 3209 to 3211 are buttons for selecting the value of the WB factor to be displayed in the radar chart 3212 from among the current value (in other words, the latest value), the value of the initial questionnaire (in other words, the value initially registered), and the average value. By referring to this factor score display area 3202, the user can confirm the balance and changes of his / her own WB factors.

[0142] The calendar 3203 indicates the user's WB level on a daily basis. The user's WB level is represented by a face mark 3213. My page button 3204 is a button for displaying my page 3200 . The recommendation button 3205 is a button for displaying a recommendation screen for WB contents (see FIG. 29).

[0143] When the user selects any date on the above calendar 3203, the My Page module 931 displays a details screen on the user terminal 802. At that time, the module generates the details screen based on the target user's activity log 1800, service action log 1400, and service information 1700. Figure 33 shows an example of the generated details screen.

[0144] The detailed screen 3300 shown in the figure has a slider 3301 indicating the WB degree, a memo display field 3302, and a selected service display field 3303. By referring to this screen, the user can check the WB degree of the selected date, the diary, and the selected WB service.

[0145] The contents of this detailed screen 3300 can be edited on an edit screen. Figure 34 shows an example of the edit screen.

[0146] The editing screen 3400 shown in the figure has a slider 3401 for specifying the WB degree, a memo input field 3402, and a record button 3403. When the record button 3403 is selected, the mypage module 931 adds the WB degree specified on this screen and the memo entered to the activity log 1800 in association with the user ID.

[0147] 2. Variations The above embodiment may be modified as follows. (1) In the above embodiment, the WB factor scores are evaluated on a 5-point scale. However, this 5-point scale is not essential. For example, the WB factor scores may be evaluated on a 2-point scale or a 10-point scale.

[0148] (2) In the above embodiment, the recommended WB content or WB service is rated on a two-level scale of "0" or "1" (see step 1911 in FIG. 19). However, this two-level evaluation is not essential. For example, the user may be asked to rate the WB content or WB service on a five-level scale, and the evaluation value may be recorded as a behavior log. The evaluation value recorded here corresponds to a reaction probability that indicates the user's reaction to the WB content or WB service.

[0149] The five-level evaluation is expressed as follows, for example: 1: Not interested at all 2: Not very interested 3: I can't decide 4: I kinda like it 5: I really liked it

[0150] (3) In the model learning in the above embodiment, the estimation model learns user attribute information (see steps 1903 and 1905 in FIG. 19). However, learning user attribute information is not essential and may be omitted.

[0151] (4) In the above embodiment, FM is used as the estimation model (see steps 1903 and 1905 in FIG. 19). However, the use of FM is not essential. A regression model other than FM may be adopted in the above embodiment.

[0152] (5) The model learning module 921 learns an individual optimal model for each individual domain. However, this learning method is not essential. The model learning module 921 may learn individual optimal models across individual domains. In this case, a single trained individual optimal model is generated.

[0153] (6) In the above recommendation process, the recommendation server 801 outputs the WB content that passed screening to the user terminal 802 (see step 1910 in FIG. 19). However, the server may output information at a stage before screening in order to have another device perform the output. For example, the server may output any one of the content behavior score information 1300, the content behavior score information 1300 reranked in step 1907, or the WB content extracted in step 1908 to another device.

[0154] Similarly, the server may output to another device, at the pre-screening stage, any of the service action score information 1500, the service action score information 1500 filtered in step 1912, the service action score information 1500 reranked in step 1913, or the WB service extracted in step 1914.

[0155] (7) In the above recommendation process, a scoring process for overall optimization is periodically performed (see step 1904 in FIG. 19). However, this scoring process may be performed when a user accesses a site.

[0156] Similarly, the scoring process for individual optimization may be executed when the user selects the WB content. In this case, the recommendation server 801 may generate the service action score information 1500 only for the WB service corresponding to the WB content selected by the user. Here, the WB service corresponding to the WB content is the WB service that shares an individual area with the WB content. In this way, the filtering process of step 1912 can be omitted.

[0157] (8) The recommendation server 801 performs both modeling and scoring on the server (steps 1903 to 1906 in FIG. 19). However, modeling is not essential. The server may cause another device to perform modeling and obtain the generated trained model from the device.

[0158] (9) In the above embodiment, the WB factors are defined in advance as latent factors by the business operator from the viewpoint of well-being. In this case, the business operator may define the WB factors in the following manner. First, the business operator defines tentative WB factors. Next, the business operator surveys users about the defined tentative WB factors. Next, the business operator performs factor analysis on the survey results to extract latent factors. Finally, the business operator defines the extracted latent factors as official WB factors. Providers may define WB factors in this manner.

[0159] (10) In the above recommendation process, when a recommended WB content is selected by a user, the factor score information 1100 of the user is updated. At that time, the updated WB factor scores may be reflected in the next modeling and scoring. Specifically, the model learning module 911 may update the trained global optimum model by training the global optimum model using the updated WB factor scores as explanatory variables. In addition, the scoring module 912 may input the updated WB factor scores into the trained global optimum model to estimate the response probability.

[0160] Similarly, the model learning module 921 may update the trained individual optimal model by training the individual optimal model using the updated WB factor scores as explanatory variables. Also, the scoring module 922 may input the updated WB factor scores to the trained individual optimal model to estimate the reaction probability.

[0161] As a result of these processes, the WB content and WB services that are next recommended to the target user may change.

[0162] The present invention is not limited to the above-described embodiments, and includes various modified examples. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0163] In addition, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0164] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. The above-described embodiments disclose at least the configurations described in the claims. [Explanation of symbols]

[0165] 800...information processing system, 801...recommendation server, 802...user terminal, 911...model learning module, 912...scoring module, 913...reranking module, 914...extraction module, 915...screening module, 916...recommendation module, 917...action log management module, 921...model learning module, 922...scoring module, 923...filtering module, 924...reranking module, 925...extraction module, 926... Screening module, 927... recommendation module, 928... behavior log management module, 931... my page module, 932... user information management module, 941... screen data, 942... image data, 1000... user attribute information, 1100... factor score information, 1200... content behavior log, 1300... content behavior score information, 1400... service behavior log, 1500... service behavior score information, 1600... content information, 1700... service information, 1800... activity log

Claims

1. A means for associating a well-being factor that expresses a value of happiness with a user based on the user's answer to a question; a first suggestion means for suggesting to the user content corresponding to the well-being factor, the content improving the well-being of the user; a first recording means for recording information about the content in association with the user when the content is selected by the user; An information processing system comprising:

2. The information processing system described in Claim 1, wherein the first suggestion means suggests content to the user that corresponds to the well-being factor and the user's interests.

3. An information processing system as described in claim 1, wherein the content is content related to sleep or eating.

4. A second suggestion means for suggesting to the user a service that improves the well-being of the user, the service being in accordance with the well-being factor; a second recording means for recording information about the service in association with the user when the service is selected by the user; The information processing system of claim 1 , further comprising:

5. The information processing system described in Claim 4, wherein the second suggestion means suggests to the user a service that corresponds to the well-being factor and the user's interests.

6. An information processing system as described in claim 4, wherein the service is a sleep-related service.

7. The information processing system described in Claim 4, wherein the second suggestion means is a service that corresponds to the well-being factor and that suggests to the user a service that corresponds to the content.

8. An information processing system as described in claim 4, further comprising a presentation means for presenting information about the selected service to the user in association with a calendar date.

9. An information processing method executed by a computer, comprising: an associating step of associating a well-being factor expressing a value of happiness level with the user based on the user's answer to the question; a first suggestion step of suggesting to the user content according to the well-being factor, the content improving the well-being of the user; a first recording step of, when the content is selected by the user, recording information of the content in association with the user; An information processing method comprising:

10. An information processing method as described in Claim 9, wherein in the first suggestion step, content corresponding to the well-being factor and the user's interests is suggested to the user.

11. An information processing method as described in Claim 9, wherein the content is content related to sleep or eating.

12. A second suggestion step of suggesting to the user a service according to the well-being factor, the service improving the well-being of the user; a second recording step of recording information about the service in association with the user when the service is selected by the user; The information processing method of claim 9 , further comprising:

13. An information processing method as described in claim 12, wherein in the second suggestion step, a service that corresponds to the well-being factor and the user's interests is suggested to the user.

14. An information processing method as described in claim 12, wherein the service is a sleep-related service.

15. An information processing method as described in claim 12, wherein in the second suggestion step, a service that corresponds to the well-being factor and that corresponds to the content is suggested to the user.

16. The information processing method according to claim 12, further comprising a presentation step of presenting information about the selected service to the user in association with a date on a calendar.

17. A program for causing a computer to function as each part described in claim 1.

Images