Information processing apparatus, information processing method, and program

Abstract

To multidirectionally support prediction of a disease onset risk of a user.SOLUTION: An information processing device acquires fact features about each of a plurality of users as user features, creates a relationship graph showing a social relationship between the plurality of users on the basis of the user features, sets a target user among the plurality of users, and predicts a disease risk showing an onset risk of one or more prescribed diseases about the target user on the basis of the relationship graph and user features of the target user.SELECTED DRAWING: Figure 8B

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program, and particularly to a technique for predicting the onset risk of a disease.

Background Art

[0002] In recent years, a technique has been developed to provide a service related to a predetermined disease to a user having a risk of onset of the disease by using a network such as the Internet. For example, Patent Document 1 discloses a technique of acquiring behavior information indicating the behavior of a user (risk user) having a risk of onset of a predetermined disease on a network, and providing information related to the predetermined disease to another user (target user) having behavior information related to the behavior information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to Patent Document 1, based on information related to the behavior of a risk user and a target user, it is estimated that the target user has the same risk of onset of the disease as the risk user, and information related to the disease is provided to the target user. However, the information related to the behavior is only local information between the risk user and the target user, and there is room for improvement in predicting the onset risk of the disease based on more comprehensive information.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a technique for more comprehensively supporting the prediction of the onset risk of a disease based on the comprehensive relationship between a user and a plurality of other users.

Means for Solving the Problem

[0006] To solve the above problems, one aspect of the information processing apparatus according to the present invention includes an acquisition means for acquiring, as user characteristics, factual characteristics for each of a plurality of users, Front a target user setting means for setting a target user among the plurality of users, and based on the Indicating the social relationships among multiple users relational graph and the user characteristics of the target user, prediction means for predicting a disease risk indicating the onset risk of one or more predetermined diseases for the target user.

[0007] The prediction means can predict the disease risk for the target user by using a machine learning model configured to input the user characteristics of the target user and output the disease risk for the target user.

[0008] It further includes learning means for training the machine learning model, and the learning means can train the machine learning model by using disease characteristics obtained from the relational graph.

[0009] The disease characteristics obtained from the relational graph may include information indicating the presence or absence of the one or more predetermined diseases for each of the plurality of users.

[0010] The relational graph In where each user is represented by each user node, and the Each user node is based on the factual characteristics Li can be connected by links Done .

[0011] In the relational graph Pairs of user nodes having the same factual characteristics Is can be connected by explicit links Done , and based on pairs of a plurality of user nodes connected by the explicit links, pairs of User nodes not connected by the explicit links Is can be connected by implicit links Done .

[0012] The relational graph is and the connected Of user nodes pair At can be based on one or more factual features shared And has an intimacy assigned to the pair thereon.

[0013] The disease risk can be represented by a numerical value from 0 to 1, with 1 being the maximum possibility, for each of the one or more predetermined diseases.

[0014] In order to solve the above problems, one aspect of the information processing method according to the present invention includes an acquisition step of acquiring factual features for each of a plurality of users as user features, Front a target user setting step of setting a target user among the plurality of users, and based on the Indicating the social relationships among multiple users relationship graph and the user features of the target user, a prediction step of predicting a disease risk indicating the onset risk of one or more predetermined diseases for the target user.

[0015] In order to solve the above problems, one aspect of the program according to the present invention is an information processing program for causing an information processing computer to execute, and the program causes the computer to perform an acquisition process of acquiring factual features for each of a plurality of users as user features, a target user setting process of setting a target user among the plurality of users, and based on the Indicating the social relationships among multiple users relationship graph and the user features of the target user, a prediction process of predicting a disease risk indicating the onset risk of one or more predetermined diseases for the target user, and is for causing the computer to execute a process including the above. [Advantages of the Invention]

[0016] According to the present invention, it becomes possible to more comprehensively support the prediction of the onset risk of a user's disease. Those skilled in the art will be able to understand the above-described objects, aspects, and effects of the present invention, as well as the objects, aspects, and effects of the present invention not described above, from the embodiments for carrying out the following invention with reference to the descriptions in the accompanying drawings and the claims.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4A

Figure 4B

Figure 4C

Figure 4D

Figure 5A

Figure 5B

Figure 6A

Figure 6B

Figure 7

Figure 8A

Figure 8B

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0018] Hereinafter, with reference to the accompanying drawings, embodiments for implementing the present invention will be described in detail. Among the components disclosed below, those having the same function are denoted by the same reference numerals, and the description thereof will be omitted. Note that the embodiments disclosed below are examples of means for realizing the present invention, and should be appropriately modified or changed according to the configuration of the apparatus to which the present invention is applied and various conditions, and the present invention is not limited to the following embodiments. Also, not all combinations of the features described in this embodiment are essential for the solution means of the present invention.

[0019] [Functional Configuration of Information Processing Apparatus] FIG. 1 shows a configuration example of an information processing system according to this embodiment. As an example, this information processing system is configured to include an information processing apparatus 10 and a plurality of user apparatuses 11-1 to 11-N (N>1) used by an arbitrary plurality of users 1 to N, as shown in FIG. 1. In the following description, unless otherwise specified, the user apparatuses 11-1 to 11-N may be collectively referred to as the user apparatus 11. Also, in the following description, the terms "user apparatus" and "user" may be used synonymously.

[0020] The user apparatus 11 is a device such as a smartphone or a tablet, and is configured to be communicable with the information processing apparatus 10 via a public network such as LTE (Long Term Evolution) or a wireless communication network such as a wireless LAN (Local Area Network). The user apparatus 11 has a display unit (display surface) such as a liquid crystal display, and each user can perform various operations through a GUI (Graphic User Interface) provided on the liquid crystal display. The operations include various operations on contents such as images displayed on the screen, such as tap operations, slide operations, and scroll operations, using a finger, a stylus, or the like. Note that the user device 11 is not limited to the device in the form shown in FIG. 1, and may be a device such as a desktop PC (Personal Computer) or a notebook PC. In that case, the operations by each user can be performed using an input device such as a mouse or a keyboard. Also, the user device 11 may separately include a display surface.

[0021] The user device 11 can log in to a web service (Internet-related service) provided from the information processing device 10 or from another device (not shown) via the information processing device 10 and use the service. The web service can include an online mall, an online supermarket, or services related to communication, finance, real estate, sports, and travel provided via the Internet. By using such a web service, the user device 11 can transmit information about the user of the user device 11 to the information processing device 10.

[0022] For example, the user device 11 can transmit information about features of the user device or the user, such as the IP (Internet Protocol) address of the user device 11, the user's address, and the user's name, to the information processing device 10. Also, the user device 11 can perform positioning calculations based on signals received from GPS (Global Positioning System) satellites (not shown) and generate the information obtained by the calculations as the position information of the user device 11 and transmit it to the information processing device 10. The information processing device 10 acquires various information from the user device 11 and creates a relational graph network (hereinafter, relational graph) indicating the social relatedness among users based on the information. Then, the information processing device 10 predicts the onset risk of any disease for any user (hereinafter, disease risk) using the created relational graph.

[0023] [Functional Configuration of Information Processing Device 10] The information processing apparatus 10 according to the present embodiment first acquires various types of information from the user devices 11-1 to 11-N, and creates a relational graph indicating the social relationships among the users 1 to N. Then, the information processing apparatus 10 identifies a target user among the users 1 to N. The information processing apparatus 10 predicts the disease risk of the target user by applying the features of the created social graph to a machine learning model whose learning has been completed.

[0024] FIG. 2 shows an example of the functional configuration of the information processing apparatus 10 according to the present embodiment. The information processing apparatus 10 shown in FIG. 2 includes a user feature acquisition unit 101, a graph creation unit 102, a target user feature setting unit 103, a prediction unit 104, a learning unit 105, an output unit 106, a learning model storage unit 110, and a user feature storage unit 120. The learning model storage unit 110 stores a disease risk prediction model 111 and a score prediction model 112. These various learning models will be described later. Also, the user feature storage unit 120 stores user features 121.

[0025] The user feature acquisition unit 101 acquires fact features (fact information) (hereinafter referred to as user features) about the user device and the user from each of the user devices 11-1 to 11-N. The user features are features (information) based on facts that can be actually or objectively obtained from the user device or the user. The user feature acquisition unit 101 can, for example, directly acquire user features from the user device 11. Also, the user feature acquisition unit 101 can acquire user features as information registered by the user of the user device 11 in a predetermined web service.

[0026] User characteristics include the IP address of the user device, the user's address, name, the number of the credit card held by the user, the user's demographic information (demographic user attributes such as gender, age, residential area, occupation, family composition, etc.), and the like. In addition, user characteristics include characteristics related to the user's health, such as health check data and daily health data. The characteristics related to health include, for example, height and weight, sleep time, dietary information (such as calorie intake), blood type, blood pressure, and past medical history, and can be obtained, for example, from data registered in a health-related service which is one of the web services.

[0027] In addition, user characteristics may include a registration number and a registration name when using a predetermined web service. Also, user characteristics may include call history, the delivery address other than the user's address for a product when using a predetermined web service, usage status when using a predetermined web service, usage history, search history, and information related to points that can be saved by using the service. Thus, user characteristics can include all information, including information related to the user device or the user himself / herself and information related to the use of a predetermined service via communication. The user characteristic acquisition unit 101 stores the acquired user characteristics in the user characteristics 120 as the user characteristics 121.

[0028] The graph creation unit 102 creates a relational graph using various user characteristics acquired by the user characteristic acquisition unit 101. The relational graph will be described later.

[0029] The target user characteristic setting unit 103 sets a user for whom disease risk prediction is to be performed (hereinafter referred to as the target user). The target user may be set by an input operation by an operator using the input unit (input unit 95 in FIG. 9), may be set in advance in the system, or may be set by an arbitrary program stored in the storage unit (ROM 92 and RAM 93 in FIG. 9). Further, the target user characteristic setting unit 103 acquires the user characteristics of the target user from the user characteristics 121 and sets them in the prediction unit 104.

[0030] The prediction unit 104 predicts a disease risk indicating the onset (acquisition) risk of one or more predetermined diseases for the target user set by the target user feature setting unit 103. In the present embodiment, the disease risk of the target user is predicted using the disease risk prediction model 111 that has been learned by the learning unit 105. The prediction process of the disease risk will be described later.

[0031] The learning unit 105 learns (trains) the disease risk prediction model 111 and the score prediction model 112, and stores the learned disease risk prediction model 111 and score prediction model 112 in the learning model storage unit 110. The learning process of each learning model will be described later.

[0032] The output unit 106 outputs the prediction result of the disease risk for the target user predicted by the prediction unit 104. The output unit 106 may generate and output information regarding the disease risk. The output may be any output process, and may be an output to an external device via the communication I / F (communication I / F 97 in FIG. 9), or may be a display on the display unit (display unit 96 in FIG. 9).

[0033] [Procedure for creating a relationship graph] Next, the procedure for creating a relationship graph according to the present embodiment will be described. In the following description, users A to E are users referred to for the purpose of explanation and may be users of the user device 11. The relationship graph is composed of the connections of each user node surrounded by a circle in FIGS. 4A to 4D. In the following description, the user node is simply referred to as a user. FIG. 3 shows a flowchart of the relationship graph creation process executed by the graph creation unit 102 according to the present embodiment. Each step of the process in FIG. 3 will be described below.

[0034] <S31: Creation of links> In S31, the graph creation unit 102 predicts and creates links between a plurality of users. The link creation process will be described with reference to FIGS. 4A to 4D. FIGS. 4A to 4C are diagrams for explaining explicit links, and FIG. 4D is a diagram for explaining implicit links. An explicit link is a link created based on an explicit common feature between two users (user pair). An implicit link is a link created as an indirect relationship using an already created explicit link, although the existence of an explicit common feature of the user pair is unclear. Thus, the links between users are identified as explicit links and implicit links.

[0035] FIG. 4A shows an example of creating an explicit link using the IP addresses of the user devices of users as a common feature. FIG. 4A shows an example where there are an online mall 41, a golf course reservation service 42, a travel-related reservation service 43, and a card management system 44 as web services available to users A to C. In FIGS. 4A to 4C, these four web services are shown, but the number of web services is not limited to a specific number.

[0036] The online mall 41 is a shopping mall available online (using the Internet). The online mall 41 can provide various products and services such as fashion, books, food, concert tickets, and real estate. The golf course reservation service 42 is operated on a website that provides online services related to golf courses and can provide, for example, golf course search and reservation and lesson information. The travel-related reservation service 43 is operated on a website that provides various travel services available online. The travel-related reservation service 43 can provide, for example, hotel and travel tour reservations, airline ticket and rental car reservations, sightseeing information, hotels, and information around hotels. The card management system 44 is operated on a website that provides services related to credit cards issued and managed by a predetermined card management company. The card management system 44 may provide services in association with at least any one of the online mall 41, the golf course reservation service 42, and the travel-related reservation service 43.

[0037] In the example of FIG. 4A, users A to C each use the online mall 41, the golf course reservation service 42, and the travel-related reservation service 43 using the same IP address (= 198.45.66.xx). Information on the IP address can be acquired by the user feature acquisition unit 101. In such a case, the graph creation unit 102 creates explicit links (for example, the link L1 between user A and user C) with each other with the feature of the same IP address as shown in the link state 45 for users A to C.

[0038] FIG. 4B shows an example of creating an explicit link using the feature of the user's address as a common feature. Similar to FIG. 4A, FIG. 4B shows an example where the online mall 41, the golf course reservation service 42, the travel-related reservation service 43, and the card management system 44 exist as web services available to users A to C. Here, users A to C each register the same address (delivery address) and use the online mall 41, the golf course reservation service 42, and the travel-related reservation service 43. Information on the address can be acquired by the user feature acquisition unit 101. In such a case, the graph creation unit 102 creates explicit links (for example, the link L1 between user A and user C) with each other with the feature of the same address as shown in the link state 46 for users A to C.

[0039] FIG. 4C shows an example of creating an explicit link using the characteristics of the credit card numbers used by the user as common characteristics. FIG. 4B, similar to FIG. 4A, shows an example where there are an online mall 41, a golf course reservation service 42, a travel-related reservation service 43, and a card management system 44 as web services available to users A to C. Here, users A to C have each registered the same credit card and are using the online mall 41, the golf course reservation service 42, and the travel-related reservation service 43. Information including the credit card number can be acquired by the user feature acquisition unit 101. In such a case, the graph creation unit 102 creates explicit links (for example, link L1 between user A and user C) with the same card characteristics among users A to C as shown in the link state 47.

[0040] FIG. 4D shows an example of creating an implicit link between users. In the example of FIG. 4D, for user A, users C, D, and E are connected by explicit links, and for user B, users C, D, and E are connected by explicit links. Such link characteristics (characteristics indicating the relationship between links) are embedded into a common feature space, and an inferred link in which an implicit relationship is constructed among each user (each node) is created (established) as an implicit link. In the example of FIG. 4D, although user A and user B are not connected by an explicit link, as a result of being inferred to have a relationship in the common feature space, an implicit link L2 is created. Note that the graph creation unit 102 predicts and creates an implicit link between users by performing learning (representation learning, relationship learning, embedding learning, knowledge graph embedding) of the user relationship graph composed of nodes (users) connected by explicit links. At this time, the graph creation unit 102 may perform the learning based on a known embedding model or its extension as appropriate.

[0041] <S32: Inference of the relationship between links> In S32, the graph creation unit 102 infers the relationship between the links predicted and created in S31. The inference process of the relationship between the links will be described with reference to FIGS. 5A and 5B. FIG. 5A is a diagram for explaining the inference process of the relationship between the links, and shows an example of inferring the relationship of the link between user A and user B connected by an explicit link.

[0042] The graph creation unit 102 treats the pair of users connected by the link created in S31 as a data point, and groups the pair (data point) into a cluster representing a common type using various information acquired by the user feature acquisition unit 101. The various information can be information such as IP address, address, credit card, age, gender, and friends. Also, each cluster can be a cluster having relationships such as spouse, parent-child, neighbor, same household, colleague, friend, same-sex siblings, opposite-sex siblings, etc. In the example of FIG. 5A, the pair of users is indicated by a cross, and as clusters into which the pair can be grouped, a parent-child cluster 51, a spouse cluster 52, a same-sex siblings cluster 53, a friend cluster 54, and a colleague cluster are shown. Note that although five clusters are shown in FIG. 5A, the number of clusters is not limited to a specific number.

[0043] For example, if user A and user B have (share) the feature 50 of having the same surname, an age difference of less than 10 years, opposite genders, and the same address, the graph creation unit 102 can group the pair of user A and user B into a cluster (spouse cluster 52) representing the relationship of husband and wife (spouse).

[0044] FIG. 5B shows a flowchart of an example of the process of grouping pairs into clusters executed by the graph creation unit 102. At the start of S51, for the pairs to be grouped, assume they have the characteristics of the same address and the same surname. In S52, the graph creation unit 102 determines whether the pairs in question have the characteristic of the same gender. If the pairs in question have the characteristic of the same gender (Yes in S52), in S53, the graph creation unit 102 determines whether the age difference between the pairs in question is less than or equal to a predetermined threshold value (= X value). If the age difference between the pairs in question is greater than the X value (No in S53), the graph creation unit 102 groups the pairs in question into the parent-child cluster 51. If the age difference is less than or equal to the X value (Yes in S53), the graph creation unit 102 groups the pairs in question into the same-gender sibling cluster 53. Also, if the pairs in question do not have the characteristic of the same gender (No in S52), in S54, the graph creation unit 102 determines whether the age difference between the pairs in question is less than or equal to a predetermined threshold value (= Y value). If the age difference is greater than the Y value (No in S54), the graph creation unit 102 groups the pairs in question into the parent-child cluster 51. If the age difference is less than or equal to the Y value (Yes in S54), the graph creation unit 102 groups the pairs in question into the spouse cluster 52.

[0045] <S33: Score Assignment Based on the Closeness of the Relationship> In S33, the graph creation unit 102 predicts a score based on the closeness of the relationship for the pairs inferred in S32 and assigns the score to the pairs. In the present embodiment, the score is a numerical value between 0 and 1, but there is no specific limitation on the numerical values that the score can take. FIG. 6A shows a conceptual diagram of the score based on the closeness of the relationship for user pairs (hereinafter referred to as the closeness score).

[0046] In the example of FIG. 6A, the intimacy of the relationship between the pair of users, which is determined by the features that users A and B have (share) and are connected by explicit links, changes. In the upper part of FIG. 6A, when users A and B have feature 60, namely being same-sex siblings, having the same address, a call history of 1,200 times, and 50 gift exchanges, the intimacy of the relationship (i.e., the intimacy score) between the pair of users is high. On the other hand, in the lower part of FIG. 6A, when users A and B have feature 61, namely being same-sex siblings, having different addresses, a call history of 30 times, and 2 gift exchanges, the intimacy of the relationship (i.e., the intimacy score) between the pair of users is low. Thus, as in the example of FIG. 6A, even for users A and B who are same-sex siblings, the intimacy of the relationship between the pair of users varies depending on other features shared by the pair of users. Pairs with high relationship intimacy are observed to have a close social distance from each other and have a high influence. On the other hand, pairs with low relationship intimacy are observed to have a far social distance from each other and not have a close relationship.

[0047] In this embodiment, the intimacy score for a pair of users is predicted using the score prediction model 112. FIG. 6B shows a schematic architecture of the score prediction model 112. The score prediction model 112 is a learning model that takes the features 63 of a pair of users as input and predicts the intimacy score 64 for the features 63.

[0048] The score prediction model 112 is, for example, a learning model that performs Weak Supervised Learning, and is, for example, a learning model based on a Convolutional Neural Network (CNN). In the present embodiment, the score prediction model 112 is a learning model that is trained using, as teacher data, the intimacy scores (0 to 1) assigned to a plurality of features for user pairs, as shown in FIG. 6A. For example, in the learning stage, as teacher data, combined data such as an intimacy score close to 1 set for the feature 60 in FIG. 6A and an intimacy score close to 0 set for the feature 61 is used. The learning process is performed by the learning unit 108. Note that the score prediction model 112 may be different for each type of relationship between user pairs, and may be a learning model trained according to one type of relationship.

[0049] In the present embodiment, the intimacy score for user pairs is predicted using the score prediction model 112. However, the graph creation unit 102 may be configured to predict the score by other methods.

[0050] Through the above processing, explicit or implicit links are formed between multiple users, an intimacy score is assigned between each link, and a relationship graph is created. A conceptual diagram of the relationship graph is shown in FIG. 7. Each of the users 71 to 73 has a plurality of features, and the intimacy score predicted as described above is assigned to user pairs.

[0051] [Prediction Process for Disease Risk] Next, the disease risk prediction process according to this embodiment will be described. In this embodiment, the prediction unit 104 uses the learned disease risk prediction model 111 to predict the disease risk for the target user. FIG. 8A shows a diagram for explaining the learning stage of the disease risk prediction model 111. The disease risk prediction model 111 takes as input the feature 81 based on the relational graph created by the graph creation unit 102, and predicts and outputs the onset (acquisition) risk (disease risk 82) of one or more predetermined diseases for each user. The disease risk prediction model 111 is, for example, a learning model based on a convolutional neural network (CNN).

[0052] In the learning stage, the learning unit 105 first prepares the feature 81, which is the input data for the disease risk prediction model 111, based on the relational graph created by the graph creation unit 102. Based on the relational graph, the learning unit 105 identifies a pair of an arbitrary user and a predetermined disease (e.g., diabetes) that the user has actually suffered from (or is suffering from; the same applies hereinafter), and assigns a positive flag to the pair. When a user has suffered from multiple diseases, a positive flag is assigned to each pair of the user and the disease. The presence or absence of the predetermined disease (the presence or absence of disease acquisition) can be determined from the medical history, etc. of each user included in the user feature 121. Further, based on the relational graph, the learning unit 105 identifies a pair of a user who has not suffered from the predetermined disease and arbitrarily set information, and assigns a negative flag to the pair. The learning unit 105 defines the pair with the positive flag as a positive data point and the pair with the negative flag as a negative data point. Note that the positive flag and the negative flag may be any values or information as long as they are distinguishable from each other. The learning unit 105 can obtain the disease characteristics of each of a plurality of users by such flag assignment from the relational graph. The disease characteristics include information indicating the presence or absence of acquisition of one or more predetermined diseases.

[0053] Subsequently, the learning unit 105 defines the personal characteristics of each user (e.g., demographic information, weight, medical history), and the characteristics between user pairs (e.g., the above-mentioned intimacy score, or one or more characteristics shared between pairs) as the input data (feature 81) of the disease risk prediction model 111. Note that the above-mentioned disease characteristics are assigned to (or included in) the personal characteristics of each user or the characteristics between the user pairs.

[0054] Using the input data (feature 81) prepared in this way, the learning unit 105 trains the disease risk prediction model 111 to output the onset (acquisition) risk (disease risk 82) of one or more predetermined diseases for each user. In this embodiment, the disease risk is represented by the possibility of each disease occurring (e.g., a numerical value from 0 to 1, where 1 indicates the highest possibility). Since the medical history of each user changes over time, each time the graph creation unit 102 updates the relational graph, the learning unit 105 can train the disease risk prediction model 111.

[0055] In the prediction stage, the prediction unit 104 predicts the disease risk for the target user. FIG. 8B shows a diagram for explaining the prediction stage of the disease risk for the target user. The prediction unit 104 inputs the user feature 83 of the target user set by the target user feature setting unit 103 into the trained disease risk prediction model 111 to predict the disease risk 84 for the target user. The disease risk 84 indicates the onset risk for one or more predetermined diseases. The user feature 83 may be any information that can identify the target user in the relational graph among the user features 121 stored in the user feature storage unit 120, and may be, for example, the address, name, demographic information, etc. of the target user.

[0056] As described above, in this embodiment, the disease risk is represented by a numerical value from 0 to 1 for each disease. When the disease risk 84 for a certain disease output from the disease risk prediction model 111 is higher than a predetermined threshold, the prediction unit 104 can determine that the onset risk of the disease is high. The predetermined threshold is, for example, 0.7.

[0057] In this way, the prediction unit 104 predicts the disease risk of the target user based on the disease risk prediction model 111 learned based on the relationship graph. Thereby, for example, it becomes possible to accurately predict the onset risk of genetic diseases caused by blood relationship (such as parent-child). Furthermore, it is also possible to accurately predict the onset risk of infectious diseases that can be infected among people living in a physically close environment, such as cohabiting family members and colleagues working in the same company.

[0058] [Hardware Configuration of Information Processing Apparatus 10] FIG. 9 is a block diagram showing an example of the hardware configuration of the information processing apparatus 10 according to the present embodiment. The information processing apparatus 10 according to the present embodiment can be implemented on a single or multiple computers, mobile devices, or any other processing platform. Referring to FIG. 9, an example in which the information processing apparatus 10 is implemented on a single computer is shown, but the information processing apparatus 10 according to the present embodiment may be implemented in a computer system including a plurality of computers. The plurality of computers may be connected to be communicable with each other via a wired or wireless network.

[0059] As shown in FIG. 9, the information processing apparatus 10 may include a CPU 91, a ROM 92, a RAM 93, an HDD 94, an input unit 95, a display unit 96, a communication I / F 97, and a system bus 98. The information processing apparatus 10 may also include an external memory. The CPU (Central Processing Unit) 91 comprehensively controls the operations in the information processing apparatus 10 and controls each component (92 to 97) via the system bus 98 which is a data transmission path.

[0060] The ROM (Read Only Memory) 92 is a non-volatile memory that stores control programs and the like necessary for the CPU 91 to execute processing. Note that the program may be stored in an external memory such as a non-volatile memory such as an HDD (Hard Disk Drive) 94, an SSD (Solid State Drive), or a removable storage medium (not shown). The RAM (Random Access Memory) 93 is a volatile memory and functions as the main memory, work area, etc. of the CPU 91. That is, when executing processing, the CPU 91 loads necessary programs and the like from the ROM 92 into the RAM 93 and realizes various functional operations by executing the programs and the like. The learning model storage unit 110 and the user feature storage unit 120 shown in FIG. 2 may be configured by the RAM 93.

[0061] The HDD 94 stores various data and various information necessary, for example, when the CPU 91 performs processing using a program. Also, the HDD 94 stores various data and various information obtained, for example, when the CPU 91 performs processing using a program or the like. The input unit 95 is composed of a pointing device such as a keyboard or a mouse. The display unit 96 is composed of a monitor such as a liquid crystal display (LCD). The display unit 86 may function as a GUI (Graphical User Interface) by being configured in combination with the input unit 95.

[0062] The communication I / F 97 is an interface that controls communication between the information processing device 10 and an external device. The communication I / F97 provides an interface with a network and executes communication with an external device via the network. Various types of data, various parameters, etc. are transmitted and received between the external device via the communication I / F97. In the present embodiment, the communication I / F87 may execute communication via a wired LAN (Local Area Network) or a dedicated line compliant with a communication standard such as Ethernet (registered trademark). However, the network available in the present embodiment is not limited to this and may be configured by a wireless network. This wireless network includes a wireless PAN (Personal Area Network) such as Bluetooth (registered trademark), ZigBee (registered trademark), and UWB (Ultra Wide Band). It also includes a wireless LAN (Local Area Network) such as Wi-Fi (Wireless Fidelity) (registered trademark) and a wireless MAN (Metropolitan Area Network) such as WiMAX (registered trademark). Furthermore, it includes a wireless WAN (Wide Area Network) such as LTE / 3G, 4G, and 5G. Note that the network may connect each device so that they can communicate with each other, and as long as communication is possible, the communication standard, scale, and configuration are not limited to the above.

[0063] Among the elements of the information processing apparatus 10 shown in FIG. 2, at least some of the functions can be realized by the CPU91 executing a program. However, at least some of the functions of the elements of the information processing apparatus 10 shown in FIG. 2 may operate as dedicated hardware. In this case, the dedicated hardware operates based on the control of the CPU91.

[0064] [Hardware Configuration of User Device 11] The hardware configuration of the user device 11 shown in FIG. 1 may be the same as that in FIG. 9. That is, the user device 10 may include a CPU 91, a ROM 92, a RAM 93, an HDD 94, an input unit 95, a display unit 96, a communication I / F 97, and a system bus 98. The user device 11 can display various information provided by the information processing device 10 on the display unit 96 and perform processing corresponding to an input operation received from the user via a GUI (configured by the input unit 95 and the display unit 96).

[0065] <Flow of processing> FIG. 10 shows a flowchart of the processing executed by the information processing device 10 according to the present embodiment. The processing shown in FIG. 9 can be realized by the CPU 91 of the information processing device 10 loading and executing a program stored in the ROM 92 or the like into the RAM 93. For the description of FIG. 10, refer to the information processing system shown in FIG. 1. It is assumed that the disease risk prediction model 111 and the score prediction model 112 that have been learned by the learning unit 105 are stored in the learning model storage unit 110.

[0066] In S101, the user feature acquisition unit 101 acquires the user features of each user from the user devices 11-1 to 11-N and stores them in the user feature storage unit 120 as user features 121. The processing in S101 may be a process of acquiring (collecting) user features in a certain past period.

[0067] In S102, the graph creation unit 102 creates a relational graph for users 1 to N using the various user features acquired by the user feature acquisition unit 101. The procedure for creating the relational graph is as described above.

[0068] In S103, the target user feature setting unit 103 sets a user (target user) who is the subject of disease risk prediction from among users 1 to N. As described above, the target user may be set by an input operation of the operator using the input unit 95, may be set in advance in the system, or may be set by any program stored in the ROM 92 or the RAM 93. Further, in S103, the target user feature setting unit 103 acquires the user features of the target user from the user features 121 and sets them in the prediction unit 104.

[0069] In S104, the prediction unit 104 inputs the user features of the target user set in S103 into the disease risk prediction model 111, and predicts a disease risk indicating the onset (acquisition) risk of one or more predetermined diseases for the target user.

[0070] In S105, the output unit 106 outputs the prediction result of the disease risk for the target user predicted in S104. The output unit 106 may generate information regarding the prediction result and output it to an external device (not shown).

[0071] For example, as in the present embodiment, the disease risk is represented by a numerical value from 0 to 1, the first threshold value is 0.1, the second threshold value is 0.3, and the third threshold value is 0.7. The output unit 106 may generate and output information from a low level to a high level according to the threshold values. For example, when the disease risk of the first disease is 0.2, the disease risk exists between the first threshold value and the second threshold value, and although it is low, it is not 0. Therefore, the output unit 106 may generate and output information indicating a warning for the first disease. Also, when the disease risk of the second disease is 0.5, the disease risk exists from the second threshold value to the third threshold value, and since the disease risk of the second disease is slightly high, the output unit 106 may generate and output information indicating the possibility of onset of the second disease. Further, when the disease risk of the third disease is 0.8, the disease risk is equal to or higher than the third threshold value, and since the disease risk of the third disease is quite high, the output unit 106 may generate and output information indicating that the onset risk of the third disease risk is high.

[0072] In this way, the information processing apparatus 10 creates a relational graph network (relational graph) indicating the social relationships among users from the user characteristics of a plurality of users, and predicts the disease risk for a target user based on the relational graph. It becomes possible to more comprehensively support the prediction of the onset risk of possible diseases for the target user (any user).

[0073] Note that although specific embodiments have been described above, the embodiments are merely illustrative and are not intended to limit the scope of the present invention. The apparatuses and methods described in this specification can be embodied in forms other than those described above. Also, without departing from the scope of the present invention, omissions, substitutions, and changes can be appropriately made to the above-described embodiments. Forms with such omissions, substitutions, and changes are included in the scope of what is described in the claims and their equivalents, and belong to the technical scope of the present invention.

Explanation of Reference Numerals

[0074] 1 to N: Users, 10: Information processing apparatus, 11-1 to 11-N: User apparatuses, 101: User characteristic acquisition unit, 102: Graph creation unit, 103: Target user characteristic setting unit, 104: Prediction unit, 105: Learning unit, 106: Output unit, 110: Learning model storage unit, 111: Disease risk prediction model, 112: Score prediction model, 120: User characteristic storage unit, 121: User characteristics

Claims

1. An acquisition means for acquiring, as user characteristics, factual characteristics including information on past medical history for each of a plurality of users; A target user setting means for setting a target user among the plurality of users; Based on the user characteristics of each arbitrary user among the plurality of users and the information included in the user characteristics of each of the one or more other users for each of the users, when the mutual information satisfies a preset condition, a learning means for training a machine learning model to output a disease risk indicating the risk of onset of each of one or more predetermined diseases for each of the users, the disease risk being a feature shared between a pair of each of the users and one or more other users; A prediction means for predicting a disease risk indicating the risk of onset of the predetermined disease for the target user by inputting the user characteristics of the target user and the features shared between the target user and one or more other users into the machine learning model trained by the learning means; An information processing apparatus, characterized by comprising the above.

2. The information processing apparatus according to claim 1, wherein the disease risk is represented by a numerical value from 0 to 1, with 1 being the maximum possibility, for each of the one or more predetermined diseases.

3. An information processing method executed by an information processing apparatus, comprising: An acquisition step of acquiring, as user characteristics, factual characteristics including information on past medical history for each of a plurality of users; A target user setting step of setting a target user among the plurality of users; A learning step of training a machine learning model to output a disease risk indicating the risk of onset of each of one or more predetermined diseases for each of the users, based on the user characteristics of each arbitrary user among the plurality of users and the information included in the user characteristics of each of the one or more other users for each of the users, the disease risk being a feature shared between a pair of each of the users and one or more other users when the mutual information satisfies a preset condition; A prediction step of predicting a disease risk indicating the risk of onset of the predetermined disease for the target user by inputting the user characteristics of the target user and the features shared between the target user and one or more other users into the machine learning model trained in the learning step; An information processing method, characterized by comprising the above.

4. An information processing program for causing a computer to execute information processing, the program causing the computer to perform an acquisition process of acquiring, as user characteristics, fact features including information on past medical history for each of a plurality of users; a target user setting process of setting a target user among the plurality of users; a learning process of training a machine learning model to output a disease risk indicating the risk of onset of each of one or more predetermined diseases for each of the users, the disease risk being identified when information between each user and one or more other users satisfies a preset condition based on the user characteristics of each of the users among the plurality of users and the information included in the user characteristics of each of the one or more other users; a prediction process of predicting a disease risk indicating the risk of onset of the predetermined disease for the target user by inputting the user characteristics of the target user and the characteristics shared between the target user and one or more other users into the machine learning model trained in the learning process, and is for causing the computer to execute a process including these processes. Information processing program.

Images