JP2019109620A - Estimation device, method for estimation, and estimation program - Google Patents

Abstract

To provide an estimation device, a method for estimation, and an estimation program that can appropriately estimate the state of a user.SOLUTION: An estimation device 100 includes an acquisition unit 131 and an estimation unit 133. The acquisition unit 131 acquires behavior information of a target user. The estimation unit 133 estimates the physical and emotional states of the target user based on the behavior information showing behaviors of the target user acquired by the acquisition unit.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an estimation apparatus, an estimation method, and an estimation program.

  Conventionally, techniques for determining or estimating various states of the user are provided. For example, there is provided a technology for determining the mood of the user based on the user's brain waves, the appearance of the imaged user, and the like.

JP, 2013-000300, A Unexamined-Japanese-Patent No. 2017-140435 JP, 2017-168054, A

  However, in the above-described prior art, it may be difficult to appropriately estimate the state of the user. For example, in the case of using the brain waves of the user, a device or the like for collecting the brain waves of the user is required, which places a heavy burden on the user. Moreover, when imaging a user, the user needs to be always positioned within the imaging range, which places a heavy burden on the user. As described above, in the above-described conventional technology, the burden on the user is large, and it is difficult to estimate the state of the user without placing a burden on the user. Further, in the above-mentioned prior art, it is necessary to arrange an apparatus for collecting brain waves and an apparatus for imaging at an appropriate place, and when these apparatuses are not arranged at an appropriate place, the state of the user It can not always be estimated properly.

  This application is made in view of the above, and an object of the present application is to provide an estimation device, an estimation method, and an estimation program which estimate a user's state appropriately.

  An estimation apparatus according to the present application estimates an object's mental and physical state based on an acquisition unit that acquires behavior information of a target user and behavior information indicating the behavior of the target user acquired by the acquisition unit. And.

  According to an aspect of the embodiment, an effect is obtained that the user's state can be appropriately estimated.

FIG. 1 is a diagram illustrating an example of estimation processing according to the embodiment. FIG. 2 is a diagram showing an example of the configuration of the estimation system according to the embodiment. FIG. 3 is a diagram illustrating an exemplary configuration of the estimation apparatus according to the embodiment. FIG. 4 is a diagram illustrating an example of the user information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of the action information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of a model information storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of the level information storage unit according to the embodiment. FIG. 8 is a diagram illustrating an example of the content information storage unit according to the embodiment. FIG. 9 is a flowchart illustrating an example of the estimation process according to the embodiment. FIG. 10 is a diagram illustrating an example of estimation processing according to a modification. FIG. 11 is a diagram illustrating an exemplary configuration of an estimation apparatus according to a modification. FIG. 12 is a diagram illustrating an example of a level estimation rule information storage unit according to a modification. FIG. 13 is a diagram illustrating an example of the anxiety action list information storage unit according to the modification. FIG. 14 is a diagram illustrating another example of the level estimation rule information storage unit according to the modification. FIG. 15 is a hardware configuration diagram showing an example of a computer that implements the function of the estimation device.

  Hereinafter, an estimation apparatus, an estimation method, and a form for implementing an estimation program (hereinafter, referred to as “embodiment”) according to the present application will be described in detail with reference to the drawings. Note that the estimation apparatus, estimation method, and estimation program according to the present application are not limited by this embodiment. Moreover, the same code | symbol is attached | subjected to the same site | part in the following each embodiment, and the overlapping description is abbreviate | omitted.

(Embodiment)
[1. Flow of estimation process]
First, an example of the estimation process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of estimation processing according to the embodiment. In the example of FIG. 1, the estimation apparatus 100 estimates a user's mental and physical state by using a learning model (hereinafter, also referred to as a “model”) generated for each combination of the user and the category. In the example of FIG. 1, it is estimated whether the user's mental and physical state is an anxiety state which is a negative state different from a normal state. Also, in the example of FIG. 1, the case of estimating whether the user's physical and mental condition is anxious will be described as an example, but the user's physical and mental condition is not limited to anxiety, and various mental and physical conditions such as peace of mind and illness. Good. Also, in the example of FIG. 1, work, health, future, unknown, etc. are shown as categories for estimating the mental and physical state, but the categories may be various categories.

[Configuration of estimation system]
First, the estimation system 1 shown in FIGS. 1 and 2 will be described. As shown in FIG. 2, the estimation system 1 includes a terminal device 10, an information provision source device 20, and an estimation device 100. The terminal device 10, the information provision source device 20, and the estimation device 100 are communicably connected by wire or wireless via a predetermined network N. FIG. 2 is a diagram showing an example of the configuration of the estimation system according to the embodiment. The estimation system 1 illustrated in FIG. 2 may include a plurality of terminal devices 10, a plurality of information provision source devices 20, and a plurality of estimation devices 100.

  The terminal device 10 is an information processing device used by a user. The terminal device 10 is realized by, for example, a smartphone, a tablet terminal, a notebook PC (Personal Computer), a desktop PC, a mobile phone, a PDA (Personal Digital Assistant), or the like. FIG. 1 shows the case where the terminal device 10 is a smartphone.

  In addition, the terminal device 10 has a function such as a GPS (Global Positioning System) sensor or the like, and can detect and obtain the position of the user. In addition, the terminal device 10 may have various functions such as a temperature sensor and an air pressure sensor, and may detect and acquire environment information on which the user is placed, such as temperature and air pressure. Further, the terminal device 10 may have various functions such as a heart rate sensor, and may be capable of detecting and acquiring biometric information of the user. For example, the user using the terminal device 10 may be able to acquire the user's own context information from the terminal device 10 by wearing a wearable device capable of communicating with the terminal device 10. For example, the user using the terminal device 10 wears a wristband type wearable device capable of communicating with the terminal device 10, whereby the terminal device 10 acquires information on the user's own heartbeat (pulse) from the terminal device 10 It may be possible.

  Also, the terminal device 10 displays the content provided from the external device. For example, the terminal device 10 transmits various information and the like to the estimation device 100. For example, the terminal device 10 transmits, to the estimation device 100, input information such as a post by a user, behavior information such as information on a search query used for a search by the user, and the like. In the following, the terminal device 10 may be described as a user. That is, the user can be read as the terminal device 10 below. The terminal device 10 receives various operations by the user.

  The estimation apparatus 100 is an information processing apparatus that estimates the mental and physical state of the target user based on the action information indicating the action of the target user. For example, the estimation apparatus 100 uses a model for each user to estimate whether the mind and body state of the target user is anxiety.

  The estimation apparatus 100 acquires user's action information from an external information processing apparatus. For example, the estimation apparatus 100 acquires user behavior information from the terminal device 10 used by the user.

  For example, the estimation apparatus 100 may provide a search service for providing the terminal device 10 with search results for a query (search query) acquired from the terminal device 10. For example, the estimation apparatus 100 may provide predetermined information corresponding to a query as a search result to the terminal device 10 that is the sender of the query. For example, the estimation device 100 may acquire the user's action information from an external information processing device or the like that provides a search service or an electronic commerce service.

  For example, the estimation apparatus 100 may acquire information posted by a user using the terminal device 10 in a predetermined service from the terminal device 10 or an external device providing the predetermined service. For example, the estimation apparatus 100 acquires, from the terminal device 10 or a predetermined external device, information posted by a user using the terminal device 10 in a social networking service (SNS) such as Twitter (registered trademark) or Facebook (registered trademark). May be For example, the estimation apparatus 100 acquires, from the terminal device 10 or an external device providing a search service, information on a search query (hereinafter, also simply referred to as a “query”) used by a user using the terminal device 10 in a predetermined search service. You may

  The information provision source device 20 is an information processing device in which information for providing various information to the estimation device 100 is stored. For example, the information providing source device 20 may store information collected from various external devices such as a web server. For example, the information provision source device 20 provides the estimation device 100 with various information such as model information and level information.

  From here, a case where the estimation apparatus 100 estimates the user U1 as a target user and estimates whether the mental state of the user U1 is anxiety will be described using FIG. 1.

  The estimation apparatus 100 acquires the action information of the user U1 from the terminal device 10 used by the user U1 (step S11). For example, the estimation device 100 acquires, from the terminal device 10, behavior information of the user U1 on the Internet. In the example of FIG. 1, the estimation apparatus 100 acquires action information ADT11 indicating that the user U1 has performed a search using the query # 11. Although illustrated as an abstract code such as "query # 11", "query # 11" may be a query related to work such as "performance slump" or "recession". Thus, the action information ADT11 includes information indicating that the user U1 has performed a search (action AC11) using the query # 11 at the date and time dt11. In the action information ADT11, the position when the user U1 performed the search using the query # 11 is the position LC11, the weather is fine, the temperature is 25 degrees Celsius (25 degrees Celsius), and the humidity is 60% Contains information to indicate that.

  In the example of FIG. 1, the estimation apparatus 100 acquires action information ADT12 indicating that the user U1 has posted in SNS # 12. In addition, although illustrated with an abstract code such as "SNS # 12", "SNS # 12" is character information or image information etc. posted on SNS such as Twitter (registered trademark), Facebook (registered trademark), etc. It is also good. Thus, the action information ADT12 includes information indicating that the user U1 has made a post (action AC12) on the SNS # 12 at the date dt12. In addition, the action information ADT12 indicates that the position when the user U1 posted in SNS # 12 is the position LC12, the weather is fine, the temperature is 25 degrees Celsius (25 degrees Celsius), and the humidity is 60%. Contains the information shown.

  The action information of the user U1 is not limited to the above, and may include action information on the Internet such as the web such as information indicating that the web content has been browsed. Also, for example, the estimation device 100 may acquire behavior information of the user U1 other than the Internet. For example, the estimation apparatus 100 may acquire various action information such as information indicating that the user U1 has purchased the B product at the A convenience store, and information indicating that the user B has eat and drink at the B restaurant.

  Then, the estimation device 100 collects action information of the user U1. In the example of FIG. 1, the estimation device 100 stores the collected action information of the user U1 in the action information storage unit 122. For example, the estimation device 100 stores the behavior information ADT11 and the behavior information ADT12 of the user U1 in the behavior information storage unit 122 in association with the user U1.

  Then, the estimation apparatus 100 selects a model corresponding to the target user from the model information storage unit 123 (step S12). Information (table) is stored in the model information storage unit 123 in FIG. 1 for each target user, such as the model data set 123-1, the model data set 123-2, the model data set 123-3, etc. Details will be described later with reference to FIG. The model information storage unit 123 in FIG. 1 stores a model data set 123-1 corresponding to the user U1, a model data set 123-2 corresponding to the user U2, and a model data set 123-3 corresponding to the user U3. Ru. Therefore, the estimation apparatus 100 selects, from the model information storage unit 123, a model data set 123-1 corresponding to the user U1 who is the target user.

  The model M1-1 illustrated in the model data set 123-1 indicates that the target category is “work (CTG1)” and is used to estimate whether the physical and mental state of the user U1 who is the target user is uneasy. Further, model data of the model M1-1 indicates that it is model data MDT1-1.

  As described above, when “model M * (* is an arbitrary character string)” is described, it indicates that the model is a model identified by the model ID “M *”. For example, when describing as "model M1-1", the model is a model identified by model ID "M1-1". In addition, as shown in a model data set 123-1 in FIG. 1, the model M1-1 is a model used to estimate whether the mind-body state in the target category "work" of the target user U1 is anxiety. , Indicates that the specific model data is "model data MDT1-1". For example, the estimation apparatus 100 indicates the degree (probability) of whether or not the mind-body state in the target category "work" of the user corresponding to the input action information is anxious by inputting the action information of the user to the model M1-1. The score is output to the model M1-1, and based on the score output by the model M1-1, the degree of anxiety of the mind-body state in the target category "work" of the user is estimated. The generation of each model such as the model M1-1 may be generated from a history of past user action information. For example, for generation of each model such as the model M1-1, learning data including action information indicating the past action of each user and correct answer information indicating that the user's mental and physical state when the action is performed indicate anxiety It may be generated. For example, generation of each model such as the model M1-1 and the like may be generated by processing as shown in the estimation device 100A shown in the modification. It may be generated from the history of past user behavior information. Further, the estimation device 100 may acquire a model used for estimation from an external device such as the information provision source device 20.

  The estimation apparatus 100 then selects a model of a category to be estimated from the model data set 123-1 (step S13). For example, the estimation apparatus 100 selects a model of a category to be estimated based on the attribute information of the user, the action content of the user, and various information related to the user. In the example of FIG. 1, since the interest of the user U1 is "economic" as shown in the user information storage unit 121, the estimation apparatus 100 estimates whether the mind-body state in the target category "work" of the user U1 is anxiety. . Therefore, in the example of FIG. 1, the estimation apparatus 100 selects the model M1-1 corresponding to the target category "job". The estimation apparatus 100 may select not only the model M1-1 but also a model such as a model M1-2 corresponding to the target category "health" or a model M1-3 corresponding to the target category "future".

  In the example of FIG. 1, the estimation apparatus 100 calculates a score indicating whether or not the mind-body state in the target category "work" of the user U1 is anxiety by the processing as shown in the processing group PS11. The estimation apparatus 100 inputs the behavior information ADT11 and ADT12 to the model M1-1 (step S14). The model M1 to which the action information ADT11 and ADT12 have been input outputs a score (step S15). In the example of FIG. 1, the model M1-1 to which the action information ADT11 and ADT12 are input outputs a score “0.85” as indicated by the score SC11.

  When the model M1-1 receives action information corresponding to one action, the estimation apparatus 100 inputs the action information ADT11 to the model M1-1 and the score output by the model M1-1 is the user U1. You may use as a score (score for estimation) which shows whether the mind-body state in the object category "job" of 5 is uneasy. Further, the estimation apparatus 100 inputs each of the action information ADT11 and ADT12 to the model M1-1, and calculates the average of the score output from the model M1-1 as to whether the mind-body state in the target category "work" of the user U1 is uneasy You may use as a score (score for estimation) shown.

  Then, the estimation apparatus 100 estimates the anxiety level of the user U1 based on the score output from the model M1-1 and the level information stored in the level information storage unit 124 (step S16).

  The level information storage unit 124 in FIG. 1 shows determination conditions based on a combination of “level” and “category”. "Level" indicates the level at which the user's mental and physical state in the corresponding category is uneasy. In FIG. 1, “level” includes items such as “low (LV1)”, “middle (LV2)”, and “high (LV3)”. “Category” indicates a category for which it is estimated whether the user's mental and physical state is uneasy. In FIG. 7, “category” includes items such as “work (CTG1)”, “health (CTG2)”, “future (CTG3)”, and “unknown (CTG4)”. The level information storage unit 124 in FIG. 1 indicates, in a box where each level and each category intersect, a determination condition as to whether the level belongs to the level according to the score output by the model.

  In the example of FIG. 1, “0.2 or more and less than 0.5” is stored in the mass at which the level “low (LV1)” and the category “work (CTG1)” intersect. That is, in the example of FIG. 1, when the category "work (CTG1)" is set as the target category, when the score output by the target user's action information is 0.2 or more and less than 0.5, the target user's anxiety in work Indicates “low (LV1)”, that is, low level. In addition, when the category “job (CTG1)” is the target category, when the score output from the action information of the target user is 0.5 or more and less than 0.8, the target user's anxiety in work is “medium (LV2) That is, the medium level is indicated. When the category “job (CTG1)” is the target category and the score output by the action information of the target user is 0.8 or more, the target user's anxiety at work is “high (LV3)”, that is, high Indicates that it is a level. The levels and the determination conditions shown in the level information storage unit 124 are an example, and the determination levels of the levels and the levels may be appropriately set according to the purpose and application.

  That is, in the example of FIG. 1, when the score output by the model M 1-1 to which the action information of the target user is input is 0.2 or more in the example of FIG. presume. In this case, the estimation apparatus 100 estimates the level of anxiety of the target user according to the score output from the model M1-1. Further, for example, when the score output by the model M1-1, to which the action information of the target user is input, is less than 0.2, the estimation apparatus 100 estimates that the mind and body state of the target user is not anxiety.

  In the example of FIG. 1, the estimation apparatus 100 has a score “0.85” output by the model M1-1 and is equal to or higher than the lower limit “0.2” of the lowest level “low (LV1)”. It is estimated that the mind-body state at work of the user U1 is anxiety. Further, as shown in the estimation list ES11, the estimation apparatus 100 has a score of “0.85”, which is output by the model M1-1, and satisfies the determination condition of “high (LV3)”. Estimate that anxiety level is high. That is, the estimation device 100 estimates that the user U1 has high anxiety at work.

  Then, the estimation apparatus 100 determines the content (provided content) to be provided to the user U1 according to the estimated anxiety level of the user U1 (step S17). In the example of FIG. 1, the estimation apparatus 100 determines the provided content to the user U1 from the content stored in the content information storage unit 125.

  In the content information storage unit 125 in FIG. 1, the A content (content CT11) identified by the content ID "CT11" indicates that the genre is sports. Further, the content CT11 indicates that the browsed user is made to have a mental and physical state of "fun". That is, the content CT11 indicates that the mental and physical state of the browsed user is brought into a positive state. For example, the content CT11 indicates that the mental and physical state of the browsed user is made safe.

  Also, the B content (content CT12) identified by the content ID “CT12” indicates that the genre is international. Further, the content CT12 indicates that the browsing user is put in a mind-body state of "sad". That is, the content CT12 indicates that the mental and physical state of the browsed user is made negative. For example, the content CT12 indicates that the mental and physical state of the browsed user is put in a state of anxiety.

  For example, since the estimation apparatus 100 estimates that the anxiety level of the user U1 is high, among the contents stored in the content information storage unit 125, the content provided to the user U1 causes the user's mental and physical state to be positive. Decide as. For example, since the estimation apparatus 100 estimates that the anxiety level of the user U1 is high, the estimation apparatus 100 determines the content that enhances the user's emotion or mood as the provided content to the user U1. In the example of FIG. 1, since the estimation apparatus 100 estimates that the anxiety level of the user U1 is high, it determines A content and C content as the provided content that brings the user into a physical and mental state of “fun”.

  Then, the estimation device 100 provides the content to the terminal device 10 used by the user U1 (step S18). For example, the estimation device 100 transmits the provided content determined in step S17 to the terminal device 10. In the example of FIG. 1, the estimation apparatus 100 distributes the A content and the C content to the terminal device 10 as the provided content.

  As described above, the estimation apparatus 100 can appropriately estimate the user's state from the user's action by estimating the user's mental and physical state based on various action information such as the user's action on the Internet. it can. In addition, the estimation apparatus 100 can appropriately estimate the state of the user from the action of the user by using a model having the action information of the user as an input. Further, the estimation apparatus 100 can appropriately estimate which level of the user's mind-body state is by using level information that determines the level of the user's mind-body state. In addition, the estimation apparatus 100 can provide the user who is uneasy about his or her physical condition with the content that makes the user's physical and mental condition be positive. Therefore, the estimation apparatus 100 can appropriately provide the user with the content according to the user's mental and physical state. Note that the estimation process may target only users who have used the user's information in the estimation process, that is, users who have opted in.

[1-1. Provided content]
In the example described above, the case where the content to be provided to the user is content such as an article corresponding to each genre is shown, but the content to be provided to the user is not limited to an article, but may be various content. For example, the content provided to the user may be an advertisement. In this case, the estimation apparatus 100 provides an advertisement to the user according to the user's mental and physical state. That is, the estimation apparatus 100 delivers an advertisement to the user according to the user's mental and physical state. For example, the estimation apparatus 100 may provide the user with an advertisement for relieving work anxiety when the mind-body state of the user's work is anxiety. For example, the estimation apparatus 100 may provide the user with an advertisement for a job change site, an advertisement for self-enlightenment, and the like when the physical and mental state of the user's work is uneasy.

  Also, for example, when the mind and body state of the user's health is anxiety, for example, the estimation device 100 may provide the user with an advertisement that relieves the health anxiety. For example, the estimation device 100 may provide the user with an advertisement for insurance, an advertisement for a gym, or the like, when the physical and mental state of the user's health is uneasy. Also, for example, when the user's physical and mental state in the future is anxiety, for example, the estimation device 100 may provide the user with an advertisement that alleviates the future anxiety. For example, when the user's physical and mental state in the future is uneasy, the estimation apparatus 100 may provide the user with an advertisement of a pension, an advertisement of qualification acquisition, and the like. Note that the above is an example, and the estimation apparatus 100 may provide various advertisements to the user according to the user's physical and mental state.

  Further, the estimation apparatus 100 may optimize the creative of the content such as the advertisement according to the estimated physical and mental state of the target user. For example, the estimation apparatus 100 may change the content of the content provided to the target user according to the estimated physical and mental state of the target user. For example, when the anxiety level of the target user estimated is high, the estimation apparatus 100 may provide the target user with an advertisement of insurance with a premium higher than normal. In the example of FIG. 1, the estimation apparatus 100 is a premium (for example, UX) in which the normal premium (for example, X yen) of A life insurance is also high because the anxiety level of the user U1 who is the target user is estimated. An advertisement ACZ of A life insurance set to circle) may be generated. In this case, the estimation apparatus 100 may include an advertisement ACZ of A life insurance having a generation unit that generates an advertisement, which indicates that the premium of A life insurance is “UX yen”. Then, the estimation device 100 may distribute an advertisement ACZ of A life insurance indicating that the premium of A life insurance is “UX yen” to the terminal device 10 of the user U1. In addition, the estimation apparatus 100 may optimize the advertisement creative according to the news (such as a sad news) reported in society (the world) as well as the user's mind and body state. For example, the estimation apparatus 100 may change the content of the advertisement to a suppressed (self-weight) content when a snail or a celebrity is reported a sad news such as an intelligence by a disease in a society to which the target user belongs.

[1-2. Mental and physical condition]
In the example described above, the case where it is estimated that the user's physical and mental state is anxiety is shown, but the estimation device 100 may estimate any user's physical and mental state as long as the user's physical and mental state. For example, the estimation device 100 may estimate whether the physical and mental state of the user is safe. In this case, the estimation apparatus 100 may perform estimation processing using a model for estimating whether the user's mental and physical state is safe. Also, for example, the estimation device 100 may estimate whether the physical and mental state of the user is in a disease state. In this case, the estimation apparatus 100 may perform the estimation process using a model for estimating whether the user's mental and physical state is a disease. Further, the mind and body state estimated by the estimation device 100 may be emotions such as joy, anger, sadness, pleasure, apathy and the like.

[1-3. category〕
In the example described above, although the case where the target category for estimating the mental and physical state is "work" is described as an example, the target categories for estimating the physical and mental state are "health", "future", "unknown" May be Further, the target category for estimating the mental and physical state may be any category as long as it is a target for estimating the physical and mental state of the user, such as "home" or "money".

[2. Configuration of estimation device]
Next, the configuration of the estimation apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram showing an exemplary configuration of the estimation apparatus 100 according to the embodiment. As shown in FIG. 3, the estimation device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The estimation apparatus 100 has an input unit (for example, a keyboard or a mouse) that receives various operations from the administrator of the estimation apparatus 100 or the like, and a display unit (for example, a liquid crystal display or the like) for displaying various information. May be

(Communication unit 110)
The communication unit 110 is realized by, for example, a network interface card (NIC). The communication unit 110 is connected to the network N in a wired or wireless manner, and transmits and receives information to and from the terminal device 10.

(Storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As illustrated in FIG. 3, the storage unit 120 according to the embodiment includes a user information storage unit 121, an action information storage unit 122, a model information storage unit 123, a level information storage unit 124, and a content information storage unit 125. Have.

(User information storage unit 121)
The user information storage unit 121 according to the embodiment stores various types of information regarding the user. For example, the user information storage unit 121 stores various information related to user attributes. FIG. 4 is a diagram illustrating an example of the user information storage unit according to the embodiment. The user information storage unit 121 illustrated in FIG. 4 includes items such as “user ID”, “age”, “sex”, “interest”, and “insurance information”. Although illustration is omitted, the user information storage unit 121 may store position information related to the user's life zone such as “home” and “working place”.

  "User ID" indicates identification information for identifying a user. "Age" indicates the age of the user identified by the user ID. The “age” may be a specific age of the user identified by the user ID, such as 35, for example. Also, “sex” indicates the gender of the user identified by the user ID.

  Also, “interest” indicates the interest of the user identified by the user ID. That is, “interesting” indicates an object that the user identified by the user ID is highly interested in. In the example shown in FIG. 4, although one “interest” is illustrated for each user, it may be plural. The “insurance information” stores information on the insurance to which the user identified by the user ID subscribes. For example, as “insurance information”, information related to various types of insurance such as the type of insurance to which the user identified by the user ID has subscribed and the insurance premium is stored. In the example shown in FIG. 4, “insurance information” is illustrated for each user, but may be plural.

  For example, in the example of FIG. 4, the age of the user (user U1) identified by the user ID “U1” is “twenties” and the gender is “male”. Also, for example, it indicates that the user U1 is interested in the economy. Also, for example, the user U1 indicates that there is no insurance subscribed.

  For example, in the example of FIG. 4, the age of the user (user U3) identified by the user ID “U3” is “30s”, and the gender is “male”. Also, for example, it indicates that the user U3 is interested in sports. Also, for example, the user U3 indicates that B insurance is enrolled and the premium is Y yen.

  The user information storage unit 121 is not limited to the above, and may store various information according to the purpose. For example, the user information storage unit 121 may store various information such as demographic attribute information and psychographic attribute information. For example, the user information storage unit 121 may store information such as name, family structure, income, and the like. For example, the user information storage unit 121 may store the position information acquired last for the user as the latest position information.

(Action information storage unit 122)
The action information storage unit 122 according to the embodiment stores various information related to the action of the user. FIG. 5 is a diagram illustrating an example of the action information storage unit according to the embodiment. For example, the action information storage unit 122 stores action information of the user such as the action of the user on the Internet. The action information storage unit 122 shown in FIG. 5 includes “user ID”, “action ID”, “date and time”, “position”, “weather (weather, temperature, humidity ...)”, “content”, “action information” Items such as are included. Moreover, although illustration is abbreviate | omitted, the action information storage part 122 memorize | stores the information (health information) which shows a user's health condition. For example, the action information storage unit 122 stores information indicating whether each user is in the normal state or in a state different from the normal state. For example, the behavior information storage unit 122 stores information indicating whether each user is in a normal state or in a negative state such as during illness or injury. For example, the behavior information storage unit 122 stores “0” when each user is in a normal state, and stores “1” as health information when in a negative state such as during illness or injury. In addition, the estimation apparatus 100 may estimate and may acquire the information which shows the health condition of each user from an external information processing apparatus. For example, the estimation device 100 may acquire information indicating the health status of each user from the information provision source device 20.

  "User ID" indicates identification information for identifying a user. For example, the user identified by the user ID “U1” corresponds to the user U1 illustrated in the example of FIG. Also, "action ID" indicates information identifying a user's action. "Date" indicates the date when the corresponding user's action was taken. Although “dt11” and the like are abstractly illustrated as “date and time”, a specific date and time such as “December 5, 2017 23:15:48” may be stored.

  Further, “position” indicates position information of the user when the action identified by the action ID is performed. In the example shown in FIG. 5, the “position” is illustrated as an abstract code such as “LC11”, but may be latitude / longitude information or the like. "Weather (weather, air temperature, humidity ...)" indicates weather information corresponding to the position of the user when the action identified by the action ID is performed. The weather information may include various information as well as weather, temperature, humidity and the like.

  Also, "content" indicates the specific content of the corresponding user's action. Also, “action information” indicates action information of the user corresponding to the action identified by the action ID. Although an example in which conceptual information such as "ADT11" is stored in "action information" is shown in FIG. 5, actually, various information related to the action taken by the user, or a file path name indicating the storage location thereof Etc. are stored. For example, the "action information" may include various information stored in the corresponding "date and time", "position", "weather (weather, temperature, humidity ...)", and "content".

  For example, in the example of FIG. 5, the user (user U1) identified by the user ID “U1” is identified by the action (action AC11) identified by the action ID “AC11” or the action identified by the action ID “AC12” Indicates that action AC12) etc. have been performed. Specifically, in the example of FIG. 5, the user U1 indicates that the search (action AC11) using the query # 11 has been performed on the date and time dt11. Further, the position when the user U1 performs the search using the query # 11 is the position LC11, which indicates that the weather is fine, the temperature is 25 degrees Celsius (25 ° C.), and the humidity is 60%. Further, the action information ADT11 indicates that the action information corresponds to the action AC11 as described above.

  The behavior information storage unit 122 is not limited to the above, and may store various information according to the purpose. Although FIG. 5 shows the case where the action information is stored in the action information storage unit 122 for each user ID, the action information is not limited to each user ID, and may be stored, for example, in time order.

(Model information storage unit 123)
The model information storage unit 123 according to the embodiment stores information on a model. For example, the model information storage unit 123 stores model information (model data) generated for each user. FIG. 6 is a diagram illustrating an example of a model information storage unit according to the embodiment. In the example of FIG. 6, the model information storage unit 123 stores information (table) for each target user, such as the model data set 123-1, the model data set 123-2, the model data set 123-3, and the like. The model data sets 123-1 to 123-3 and the like illustrated in FIG. 6 include items such as “target user”, “model ID”, “target category”, and “model data”. Although the models M1-1 to M3-4 and the like are illustrated in FIG. 6, a large number of pieces of model information may be stored according to the combination of each user and category, such as M1-5 and M3-5.

  In the example of FIG. 6, in the model data set 123-1, a model group corresponding to the user (user U1) identified by the user ID “U1” who is the target user is stored. The model data set 123-2 stores a group of models corresponding to the user (user U2) identified by the user ID "U2" who is the target user. The model data set 123-3 stores a model group corresponding to the user (user U3) identified by the user ID "U3" who is the target user.

  "Model ID" indicates identification information for identifying a model. For example, the model identified by the model ID “M1-1” corresponds to the model M1-1 illustrated in the example of FIG. The “target category” indicates the target category of the corresponding model. Also, “model data” indicates data of the corresponding model. For example, “model data” includes information including nodes in each layer, functions adopted by each node, connection relationships between nodes, and connection coefficients set for connections between nodes.

  For example, in the example illustrated in FIG. 6, the model (model M1-1) identified by the model ID "M1-1" has a target category "work (CTG1)", and the mind-body state of the target user's work is uneasy Indicates that it is used to estimate whether or not. Further, model data of the model M1-1 indicates that it is model data MDT1-1.

  For example, the model M1-1 (model data MDT1-1) is any layer from the input layer to which the user's action information is input, the output layer, and the input layer to the output layer, and is a layer other than the output layer And the second element whose value is calculated based on the first element and the weight of the first element, and the action information input to the input layer is included in each layer other than the output layer. The computer functions so that the value of the score used to estimate the user's mental and physical state is output from the output layer by performing an operation based on the first element and the weight of the first element with each element to which it belongs as the first element. It is a model to make

Here, the model M1-1~M3-4 like are realized by the regression model indicated by _{"y = a 1 * x 1 +} a 2 * x 2 + ··· + a i * x i ." In this case, for example, the first element included in the model M1-1 corresponds to input data (xi) such as x1 and x2. Also, the weight of the first element corresponds to the coefficient ai corresponding to xi. Here, the regression model can be regarded as a simple perceptron having an input layer and an output layer. When each model is regarded as a simple perceptron, the first element can correspond to any node of the input layer, and the second element can be regarded as a node of the output layer.

  Further, it is assumed that the models M1-1 to M3-4 and the like are realized by a neural network such as DNN (Deep Neural Network) having one or more intermediate layers. In this case, for example, the first element included in the model M1-1 corresponds to any node that the input layer or the middle layer has. Also, the second element corresponds to the next-stage node that is a node to which a value is transmitted from the node corresponding to the first element. Also, the weight of the first element corresponds to the connection coefficient, which is a weight considered for the value transmitted from the node corresponding to the first element to the node corresponding to the second element.

  The model information storage unit 123 is not limited to the above, and may store various model information according to the purpose. For example, the model information storage unit 123 may store a model that is generated for each user attribute and that is used to estimate the physical and mental state of the user corresponding to the user attribute. For example, the model information storage unit 123 is a model generated for each user attribute such as a 20's male, a 20's female, a 30's male, etc., and a model used to estimate the mental and physical state of the user corresponding to the user attribute. May be stored.

(Level information storage unit 124)
The level information storage unit 124 according to the embodiment stores various information related to the level. For example, the level information storage unit 124 stores various information regarding the level of each category. FIG. 7 is a diagram illustrating an example of the level information storage unit according to the embodiment. For example, the level information storage unit 124 stores determination conditions based on a combination of “level” and “category”. Specifically, the level information storage unit 124 stores information indicating determination conditions based on the score of the user's mental and physical state level in the corresponding category in each of the cells in the row “level” and the column “category”. Be done. In the example of FIG. 7, the level information storage unit 124 stores information indicating determination conditions based on the score of the level of anxiety of the user in the corresponding category in each box of row “level” and column “category”. Be done.

  "Level" indicates the level of the user's mental and physical state in the corresponding category. In FIG. 7, “level” includes items such as “low (LV1)”, “middle (LV2)”, and “high (LV3)”. Note that the above is an example, and the “level” is not limited to three levels of low, medium, and high, and may be various stages (for example, ten stages).

  “Category” indicates a category for which the user's mental and physical state is to be estimated. In FIG. 7, “category” includes items such as “work (CTG1)”, “health (CTG2)”, “future (CTG3)”, and “unknown (CTG4)”. Note that the above is an example, and any category may be used as the "category" as long as the category is a target for estimating the physical and mental state of the user, such as "home" and "money".

  In the example of FIG. 7, the squares at which each level and each category intersect indicate a determination condition as to whether the level belongs to the level corresponding to the score output by the model.

  In the example of FIG. 7, “0.2 or more and less than 0.5” is stored in the mass at which the level “low (LV1)” and the category “work (CTG1)” intersect. That is, in the example of FIG. 7, when the category “job (CTG1)” is the target category, the score output by the action information of the target user is 0.2 or more and less than 0.5 (for example, 0.35). Anxiety of the target user at work indicates "low (LV1)", that is, low level. When the category “job (CTG1)” is the target category, if the score output by the action information of the target user is 0.5 or more and less than 0.8 (for example, 0.7 or the like), Anxiety indicates "Medium (LV2)", or medium level. When the category “job (CTG1)” is the target category, the target user's anxiety at work is “high” when the score output by the target user's action information is 0.8 or more (for example, 0.9) (LV3) ", which indicates high level.

  The level information storage unit 124 may store various information according to the purpose, not limited to the above.

(Content information storage unit 125)
The content information storage unit 125 according to the embodiment stores various information related to content (articles). FIG. 8 is a diagram illustrating an example of the content information storage unit according to the embodiment. The content information storage unit 125 illustrated in FIG. 8 has items such as “content ID”, “content”, and “content”.

  "Content ID" indicates identification information for identifying content. "Content" indicates content acquired from a provider such as a so-called content provider. Although FIG. 8 shows an example in which conceptual information such as "A content" is stored in "content", actually, character information or a combination of character information and an image, or a file indicating the storage location of these Path name etc. are stored.

  Also, "content" indicates the specific content of the corresponding content. For example, "content" indicates the genre (type) of the corresponding content. Also, “content” stores information indicating a psychological action to be given to the user who browsed the content. The parentheses in the “content” shown in FIG. 8 indicate what kind of mind and body the user who browsed the content becomes.

  In the example of FIG. 8, the A content (content CT11) identified by the content ID "CT11" indicates that the genre is sports. Further, the content CT11 indicates that the browsed user is made to have a mental and physical state of "fun". That is, the content CT11 indicates that the mental and physical state of the browsed user is brought into a positive state. For example, the content CT11 indicates that the mental and physical state of the browsed user is made safe.

  Also, the B content (content CT12) identified by the content ID “CT12” indicates that the genre is international. Further, the content CT12 indicates that the browsing user is put in a mind-body state of "sad". That is, the content CT12 indicates that the mental and physical state of the browsed user is made negative. For example, the content CT12 indicates that the mental and physical state of the browsed user is put in a state of anxiety.

  The content information storage unit 125 is not limited to the above, and may store various information according to the purpose. For example, the content information storage unit 125 may store information regarding the date and time when the content was acquired and the date and time when the content was created. The content information storage unit 125 may store information on the evaluation value of each provider. In addition, the content information storage unit 125 may store identification information for identifying a content providing source.

(Control unit 130)
Returning to the description of FIG. 3, the control unit 130 is a controller, and is stored in a storage device inside the estimation apparatus 100 by, for example, a central processing unit (CPU) or a micro processing unit (MPU). Various programs (corresponding to an example of a distribution program) are realized by executing the RAM as a work area. The control unit 130 is a controller, and is realized by, for example, an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The control unit 130 controls the input layer to which the user's action information is input, the output layer, and the input layer to the output layer by the information processing according to the model M1-1 and the like stored in the model information storage unit 123. Is input to the input layer, including a first element belonging to any layer but a layer other than the output layer, and a second element whose value is calculated based on the first element and the weight of the first element By performing an operation based on the first element and the weight of the first element with each element belonging to each layer other than the output layer as the first element, the score of the score used for estimating the mental and physical state of the user Output value from output layer.

  As shown in FIG. 3, the control unit 130 includes an acquisition unit 131, a selection unit 132, an estimation unit 133, a selection unit 134, a determination unit 135, and a provision unit 136, and the information described below Implement or execute the processing function or action. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 3, and may be another configuration as long as it performs the information processing described later. For example, when the estimation device 100 generates a model, the control unit 130 may include the generation unit 136 of the estimation device 100A according to the modification. Also, in this case, the storage unit 120 may include the level estimation rule information storage unit 123A and the anxiety action list information storage unit 124A. For example, when the estimation device 100 generates a model, the estimation device 100 may be integral with the estimation device 100A according to the modification.

  The control unit 130 calculates a value based on the first element and the weight of the first element by information processing according to the model M1-1 (model data MDT1-1) stored in the storage unit 120. By performing an operation based on the first element and the weight of the first element, with each element belonging to each layer other than the output layer being the first element, including two elements, with respect to the behavior information input to the input layer , The value of the score used to estimate the user's mental and physical state is output from the output layer. The model such as the model M1-1 may be assumed to be used as a program module which is a part of artificial intelligence software. The model such as the model M1-1 may be a program.

(Acquisition unit 131)
The acquisition unit 131 acquires various information. For example, the acquisition unit 131 acquires an information request from an external information processing apparatus. For example, the acquisition unit 131 acquires various types of information from an external device such as the terminal device 10. Further, the acquisition unit 131 acquires various types of information from the storage unit 120. Further, the acquisition unit 131 acquires various types of information from the user information storage unit 121, the action information storage unit 122, the model information storage unit 123, the level information storage unit 124, the content information storage unit 125, and the like. Further, the acquisition unit 131 acquires information according to the selection of the selection unit 132. In addition, the acquisition unit 131 acquires information according to the estimation of the estimation unit 133. In addition, the acquisition unit 131 acquires information indicating the health condition of the target user. The acquisition unit 131 acquires information indicating the health condition of the target user from the user information storage unit 121. For example, the acquisition unit 131 acquires information indicating the health condition of the target user from an external information processing apparatus. In addition, the acquisition unit 131 acquires information according to the determination of the determination unit 135.

  The acquisition unit 131 acquires the action information of the target user. In addition, the acquisition unit 131 acquires behavior information of the target user on the Internet. Further, the acquisition unit 131 acquires behavior information including query information used by the target user for a search. In addition, the acquisition unit 131 acquires behavior information including purchase information of the target user. In addition, the acquisition unit 131 acquires behavior information including position information of the target user. In addition, the acquisition unit 131 acquires behavior information including weather information at a position corresponding to the target user.

  In addition, the acquisition unit 131 acquires a model used to estimate the mental and physical state of the user from the action information of the user. In addition, the acquisition unit 131 acquires a model used to estimate which of the plurality of levels of the mind and body state the mind and body state of the user is. The acquisition unit 131 also acquires a plurality of models generated for each user. In addition, the acquisition unit 131 acquires a plurality of models generated for each of a plurality of categories which are targets for estimating the mental and physical state of the user.

  The acquisition unit 131 acquires user's action information. For example, the acquisition unit 131 acquires the action information of the user from the terminal device 10 used by the user. For example, the acquisition unit 131 acquires behavior information of a target user who is a content provision target.

  In the example of FIG. 1, the acquisition unit 131 acquires behavior information of the user U1 from the terminal device 10 used by the user U1. For example, the acquisition unit 131 acquires, from the terminal device 10, behavior information of the user U1 on the Internet. For example, the acquisition unit 131 acquires behavior information ADT11 indicating that the user U1 has performed a search using the query # 11. For example, the acquisition unit 131 acquires the action information ADT12 indicating that the user U1 has made a post in the SNS # 12.

(Selection unit 132)
The selection unit 132 selects various pieces of information. For example, the selection unit 132 selects a model used for estimation of the mind-body state.

  The selection unit 132 selects one model based on a predetermined condition among a plurality of models generated for each predetermined object. The selection unit 132 selects one model corresponding to the target user among the plurality of models generated for each predetermined target. The selection unit 132 selects one of the plurality of models that corresponds to a target category to be estimated.

  In the example of FIG. 1, the selection unit 132 selects a model corresponding to the target user from the model information storage unit 123. The selection unit 132 selects, from the model information storage unit 123, a model data set 123-1 corresponding to the user U1 who is the target user. Further, the selection unit 132 selects a model of a category to be estimated from the model data set 123-1. The selection unit 132 selects a model of a category to be estimated based on the user attribute information, the user's action content, and various information related to the user. In the example of FIG. 1, as shown in the user information storage unit 121, since the interest of the user U1 is "economic", the selection unit 132 estimates whether the mind-body state in the target category "work" of the user U1 is anxiety. . The selection unit 132 selects the model M1-1 corresponding to the target category “task”.

(Estimation unit 133)
The estimation unit 133 estimates various information. For example, the estimation unit 133 estimates the mental and physical state of the user using the information acquired by the acquisition unit 131. For example, the estimation unit 133 estimates the mental and physical state of the user.

  The estimation unit 133 estimates the mental and physical state of the target user based on the action information indicating the action of the target user acquired by the acquisition unit 131. Further, the estimation unit 133 estimates whether the physical and mental state of the target user is another state different from the normal state. Further, the estimation unit 133 estimates whether or not the mental and physical state of the target user is a negative state different from the normal state. Further, the estimation unit 133 estimates whether or not the mind and body state of the target user is uneasy.

  The estimation unit 133 estimates the mental and physical state of the target user based on the target user's behavior information and the model. The estimation unit 133 estimates the level of the mental and physical state of the target user based on the score output by the model when the target user's action information is input. The estimation unit 133 estimates the mental and physical state of the target user based on the one model selected by the selection unit 132. The estimation unit 133 estimates the mental and physical state of the target user based on the one model selected by the selection unit 132. The estimation unit 133 estimates the mental and physical state of the target user in the target category based on the one model selected by the selection unit 132.

  In the example of FIG. 1, since the interest of the user U1 is "economic" as shown in the user information storage unit 121, the estimation unit 133 estimates whether the mind-body state in the target category "work" of the user U1 is uneasy . The estimation unit 133 estimates the anxiety level of the user U1 based on the score output from the model M1-1 and the level information stored in the level information storage unit 124. When the score output by the model M1-1 to which the action information of the target user is input is 0.2 or more, the estimation unit 133 estimates that the mind-body state of the target user is uneasy. The estimation unit 133 estimates the level of anxiety of the target user according to the score output from the model M1-1. In addition, for example, when the score output by the model M1-1 to which the action information of the target user is input is less than 0.2, the estimation unit 133 estimates that the mind and body state of the target user is not anxiety.

  In the example of FIG. 1, the estimation unit 133 has a score “0.85” output by the model M1-1 and is equal to or higher than the lower limit “0.2” of the lowest level “low (LV1)”. It is estimated that the mind-body state at work of the user U1 is anxiety. In addition, as shown in the estimation list ES11, the estimation unit 133 has a score of “0.85” output from the model M1-1, and satisfies the determination condition of “high (LV3)”. Estimate that anxiety level is high. For example, the estimation unit 133 estimates that the user U1 has high anxiety at work.

(Selection unit 134)
The selection unit 134 selects various pieces of information. For example, the selection unit 134 selects a user (processing target user) who is a target of a predetermined process, based on information indicating the user's mental and physical state and information indicating the user's health state.

  For example, the selection unit 134 selects the processing target user who is the user who is the target of the predetermined process based on the information indicating the user's health condition and the mental and physical condition of the user estimated by the estimation unit 133. . The selection unit 134 selects the target user as the processing target user when the difference between the health state of the target user and the physical and mental state of the target user satisfies a predetermined condition. The selection unit 134 selects the target user as the processing target user when the health state of the target user is a normal state and the physical and mental state of the target user is another state different from the normal state.

  For example, the selecting unit 134 selects the target user as the processing target user when the health state of the target user is a normal state and the mental and physical state of the target user is a negative state. For example, when the health condition of the target user is the normal state and the mind and body state of the target user is the uneasy state, the selection unit 134 selects the target user as the processing target user.

  In the example of FIG. 1, the selection unit 134 determines that the mental state of the user U1 is negative because the anxiety level of the work of the user U1 who is the target user is "high", according to the health state of the user U1. The user U1 may be selected as the process target user. For example, the selection unit 134 may select the user U1 as the processing target user based on the health condition of the user U1 stored in the user information storage unit 121. For example, when the health condition of the user U1 is normal, the selection unit 134 may select the user U1 as the processing target user. Further, for example, when the health condition of the user U1 is a negative state different from the normal state, the selection unit 134 may not select the user U1 as the processing target user. As described above, when there is a gap between the user's mental state and the health state, the estimation apparatus 100 selects the user as the process target target more efficiently by selecting the user as the process target user. be able to.

(Determination unit 135)
The determination unit 135 determines various information. For example, the determination unit 135 determines the content to be distributed to the terminal device 10. For example, the determination unit 135 determines the service to be provided to the process target user selected by the selection unit 134. For example, the determination unit 135 determines to provide various services to the processing target user selected by the selection unit 134. For example, the determination unit 135 determines to provide various services such as insurance sales services and insurance sales services to the processing target user selected by the selection unit 134.

  For example, the determination unit 135 determines the premium to be applied to the target user based on the physical and mental state of the target user estimated by the estimation unit 133. The determination unit 135 determines the content to be distributed to the target user based on the physical and mental state of the target user estimated by the estimation unit 133. The determination unit 135 determines the content to be provided to the target user based on the physical and mental state of the target user estimated by the estimation unit 133.

  In the example of FIG. 1, the determination unit 135 determines the content (provided content) to be provided to the user U1 in accordance with the estimated anxiety level of the user U1. From the content stored in the content information storage unit 125, the determination unit 135 determines the provided content to the user U1. Since the determination unit 135 estimates that the anxiety level of the user U1 is high by the estimation unit 133, among the contents stored in the content information storage unit 125, the content for making the user's mental and physical state positive is set to the user U1. Determined as provided content of Since the determination unit 135 estimates that the anxiety level of the user U1 is high by the estimation unit 133, the determination unit 135 determines, as the provided content to the user U1, the content that enhances the user's emotions and moods. Since the determination unit 135 estimates that the anxiety level of the user U1 is high by the estimation unit 133, the determination unit 135 determines A content and C content that cause the user to have a physical and mental state of “fun” as the provided content.

(Provision unit 136)
The providing unit 136 provides various information. For example, the providing unit 136 provides various information to external devices such as the terminal device 10 and the information providing source device 20. For example, the providing unit 136 transmits various information to an external device such as the terminal device 10. For example, the providing unit 136 distributes various information to an external device such as the terminal device 10. For example, the providing unit 136 provides the terminal device 10 with the provided content determined by the determining unit 135. For example, the providing unit 136 provides various services to the processing target user selected by the selecting unit 134. For example, the providing unit 136 provides the service determined by the determining unit 135 to the processing target user selected by the selecting unit 134. For example, the providing unit 136 provides various services such as insurance sales services and insurance sales services to the processing target user selected by the selection unit 134. For example, the providing unit 136 performs various information provision services such as content provision on the processing target user selected by the selection unit 134.

  For example, the providing unit 136 provides a service based on the physical and mental state of the target user estimated by the estimating unit 133. In addition, the providing unit 136 provides an information providing service based on the physical and mental state of the target user estimated by the estimating unit 133. The providing unit 136 also provides an advertisement distribution service based on the physical and mental state of the target user estimated by the estimating unit 133.

  In the example of FIG. 1, the providing unit 136 provides the content to the terminal device 10 used by the user U1. Also, the providing unit 136 transmits the provided content determined by the determining unit 135 to the terminal device 10. In the example of FIG. 1, the providing unit 136 distributes the A content and the C content to the terminal device 10 as the provided content. For example, when the user U1 is selected as the process target user by the selection unit 134, the providing unit 136 may provide information to the terminal device 10-1. Further, for example, when the user U1 is not selected as the process target user by the selection unit 134, the providing unit 136 may not provide the information to the terminal device 10-1.

[3. Flow of estimation process]
Next, the procedure of the estimation process by the estimation system 1 according to the embodiment will be described using FIG. 9. FIG. 9 is a flowchart illustrating an example of the estimation process according to the embodiment.

  First, estimation processing of tag information to be associated with content will be described with reference to FIG. As shown in FIG. 9, the estimation apparatus 100 acquires a model used for estimation of the user's mental and physical state (step S101). For example, the estimation apparatus 100 acquires a model corresponding to a target user. In the example of FIG. 1, the estimation apparatus 100 selects and acquires a model corresponding to a target user from the model information storage unit 123.

  In addition, the estimation device 100 acquires action information of the target user (step S102). For example, the estimation apparatus 100 acquires action information of the user U1. In the example of FIG. 1, the estimation device 100 acquires the action information ADT11, the action information ADT12, etc. of the user U1.

  Then, the estimation apparatus 100 estimates the mental and physical state of the target user based on the target user's action information and the model (step S103). For example, the estimation apparatus 100 estimates the mental and physical state of the target user based on the score output by the model by inputting the target user's action information to the model. In the example of FIG. 1, the estimation apparatus 100 inputs the action information ADT11, the action information ADT12, and the like of the user U1 to the model M1-1 corresponding to the category "work", and according to the score output by the model M1-1. The anxiety level of the user U1 in the category "work" is estimated.

[4. Modified example]
The estimation system 1 according to the above-described embodiment may be implemented in various different modes other than the above-described embodiment. For example, estimation processing using a predetermined list of anxiety behaviors and level estimation rules may be performed. So, below, other embodiment of said estimation system 1 is described. In addition, about the point similar to embodiment mentioned above, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably. For example, the estimation system 1 according to the embodiment and the estimation system in the modification have the same configuration except for the difference between the estimation apparatus 100 and the estimation apparatus 100A. Hereinafter, the estimation process by the estimation device 100A will be described.

[4-1. Estimation process]
First, estimation processing in the estimation system according to the modification will be described with reference to FIG. FIG. 10 is a diagram illustrating an example of estimation processing according to a modification. The case where estimation device 100A estimates user U2 as a target user and estimates whether the mental state of user U2 is anxiety will be described using FIG.

  The estimation device 100A acquires the action information of the user U2 from the terminal device 10 used by the user U2 (step S21). For example, the estimation device 100A acquires action information of the user U2 in a predetermined period (for example, one week). For example, the estimation device 100A acquires the action information ADT21 and ADT22 of the user U2 of the latest one week.

  For example, the estimation device 100A acquires, from the terminal device 10, behavior information of the user U2 on the Internet. In the example of FIG. 10, the estimation device 100A acquires behavior information ADT21 indicating that the user U2 has browsed the content # 21. In addition, although illustrated by an abstract code such as "content # 21", "content # 21" is a "DS site" which is a site related to a predetermined disease or an "MS site" which is a site related to a predetermined medicine. May be Thus, the action information ADT 21 includes information indicating that the user U 2 has performed browsing of the content # 21 (action AC 21) on the date and time dt 21. In the action information ADT21, the position at which the user U2 performed a search using the content # 21 was the position LC21, the weather was rain, the temperature was 27 degrees Celsius (27 degrees Celsius), and the humidity was 90% Contains information to indicate that.

  In the example of FIG. 10, the estimation device 100A acquires behavior information ADT22 indicating that the user U2 has purchased (purchased) the product # 22. Although illustrated with an abstract code such as "product # 22", "product # 22" may be "P medicine" which is a sleep inducing agent, "F book" which is a book of fortune telling, or the like. Thus, the action information ADT 22 includes information indicating that the user U 2 has purchased the item # 22 (action AC 22) at the date and time dt 22. Further, in the action information ADT22, the position when the user U2 purchases the product # 22 is the position LC22, the weather is cloudy, the temperature is 29 degrees Celsius (29 degrees Celsius), and the humidity is 80%. Contains information indicating

  Then, the estimation device 100A collects action information of the user U2. In the example of FIG. 10, the estimation device 100A stores the collected action information of the user U2 in the action information storage unit 122. For example, the estimation device 100A stores the behavior information ADT 21 and the behavior information ADT 22 of the user U 2 in the behavior information storage unit 122 in association with the user U 2.

  Then, the estimation device 100A estimates the anxiety level of the user U2 for each category (step S22). In the example of FIG. 10, the estimation device 100A estimates the anxiety level of the user U2 using the information stored in the level estimation rule information storage unit 123A and the anxiety action list information storage unit 124A.

  The anxiety action list information storage unit 124A in FIG. 10 stores list information of anxiety actions associated with each category. The anxiety action list information storage unit 124A stores information on actions counted (counted) as anxiety actions for each category. The details of the anxiety action list information storage unit 124A will be described in detail with reference to FIG.

  For example, the estimation apparatus 100A counts, for each category, the number of anxiety behaviors included in the action information ADT21, ADT22, etc. of the user U2 in a predetermined period. For example, the estimation device 100A counts, for each category, the number of anxiety behaviors included in the action information ADT21, ADT22, and the like of the user U2 for the latest one week. Although the illustration is omitted in the example of FIG. 10, in order to simplify the explanation, the number of times of anxiety behavior corresponding to the category "health" included in the action information ADT21, ADT22, etc. of the user U2 for the last week is "12 The case where the number of anxiety behaviors corresponding to the categories “work”, “future”, and “unknown” is “0” is described.

  For example, the estimation apparatus 100A estimates the anxiety level of the user U2 in each category using the counted number of times of anxiety behavior of each category and the information in the level estimation rule information storage unit 123A.

  The level estimation rule information storage unit 123A in FIG. 10 has information indicating a rule that is based on the score of the user's mental and physical state level in the corresponding category for each row that is defined as row "level" and column "category" (rule (rule Information). "Level" indicates the level of the user's mental and physical state in the corresponding category. In FIG. 10, “level” includes items such as “low (LV1)”, “middle (LV2)”, and “high (LV3)”.

  “Category” indicates a category for which the user's mental and physical state is to be estimated. In FIG. 10, the “category” includes items such as “work (CTG1)”, “health (CTG2)”, “future (CTG3)”, and “unknown (CTG4)”.

  For example, in the box where each level and each category intersect, a level estimation rule (level estimation model) for estimating whether the user's mental and physical state exists in each level in the corresponding category is indicated. In the example of FIG. 10, a level estimation rule for estimating whether the user's anxiety level falls under each level in the corresponding category is shown in the box in which each level and each category intersect.

  In the example of FIG. 10, “the number of times of anxiety behavior for a predetermined period is 10 times or more” is stored in the box where the level “high (LV3)” and the category “health (CTG2)” intersect. That is, in the example of FIG. 10, in the category “health (CTG 2)”, when the number of times of anxiety behavior included in the action information of the target user in a predetermined period (for example, one week) is 10 or more, Anxiety indicates "high (LV3)" or high level.

  For example, the estimation apparatus 100A determines the number of times of anxiety behavior (hereinafter also referred to as “the number of times of anxiety”) included in the action information of the target user for the last week and the threshold of each level (three times, five times, ten times, etc.) Compare with. For example, when the number of anxiety times is equal to or higher than the high level threshold “10”, the estimation device 100A estimates that the anxiety of the target user's health is at a high level. Also, for example, when the number of anxiety is less than the high level threshold “10” and equal to or higher than the middle level threshold “5”, the estimation device 100A estimates that the health user's anxiety is at the middle level. Further, for example, when the number of anxiety is less than the middle level threshold “5” and the low level threshold “3”, for example, the estimation device 100A estimates that the anxiety of the target user's health is at the low level.

  In the example of FIG. 10, the estimation device 100A determines that the number of times of anxiety behavior corresponding to the category "health" is "12" times, and the "predetermined period of time" of the level "high (LV3)" of the category "health (CTG2)". It is determined that the condition of “the number of anxiety behaviors is 10 or more” is satisfied. Therefore, the estimation device 100A estimates that the anxiety level of the user U2's health is high, as shown in the estimated level list UL21.

  Further, in the example of FIG. 10, since the number of anxiety behaviors corresponding to the categories “work”, “future”, and “unknown” is “0” times in the estimation device 100A, the work of the user U2, the future, and the unknown It is estimated that there is no anxiety. Therefore, as shown in the estimated level list UL21, the estimation apparatus 100A estimates the work, future, and unknown anxiety levels of the user U2 as "none (LV0)."

  Then, the estimation apparatus 100A determines the premium of the user U2 in accordance with the estimated anxiety level of the user U2 (step S23). In the example of FIG. 10, as shown in the user information storage unit 121, the estimation apparatus 100A recommends the user U2 to join according to the estimated anxiety level of the user U2 because the user U2 has not subscribed to insurance. Determine the premium to be presented.

  For example, since the estimation device 100A estimates that the anxiety level of the user U2 is high, the estimation device 100A determines to present the user U2 a premium higher than the premium paid by the other users. For example, the estimation apparatus 100A determines that the user U2 is to be provided with information to recommend subscription to insurance subscribed by a user whose user attribute is similar to that of the user U2. Further, since the estimation device 100A estimates that the anxiety level of health of the user U2 is high, the estimation device 100A determines to present the user U2 a premium higher than the premium paid by the user whose user attribute is similar to that of the user U2.

  In the example of FIG. 10, the estimation apparatus 100A determines to present the C cancer insurance to which the user U5 who is a 20-year-old woman is subscribed to the user U2 at a premium higher than the premium paid by the user U5. For example, the estimation apparatus 100A determines to recommend the C cancer insurance to the user U2 with the UZ yen higher than the premium Z yen paid by the user U5 as the premium.

  Then, the estimation device 100A provides information to the terminal device 10 used by the user U2 (step S24). For example, the estimation device 100A transmits, to the terminal device 10 used by the user U2, information for inviting participation in C cancer insurance. In the example of FIG. 10, the estimation device 100A is a terminal device 10 in which the user U2 uses an advertisement (for example, an advertisement ACX) that recommends C cancer insurance with a UZ yen higher than the premium Z yen paid by the user U5 as a premium. Deliver to. For example, the estimation device 100A distributes the advertisement ACX including the information indicating C cancer insurance and the information indicating that the premium is UZ yen to the terminal device 10 that the user U2 uses.

  As described above, the estimation device 100A appropriately estimates the user's state from the user's action by estimating the user's mental and physical state based on various action information such as the user's action on the Internet. it can. In addition, the estimation device 100A can appropriately estimate the state of the user from the action of the user by using the number of anxiety actions included in the action information of the user and the level estimation rule. In addition, the estimation device 100A can provide, to a user whose mind and body state is anxiety, information corresponding to an object (category) in which the user feels anxiety. For example, the estimation device 100A can appropriately provide the user with information such as an advertisement according to the user's physical and mental state.

[4-2. Configuration of estimation device]
Next, the configuration of an estimation apparatus 100A according to a modification will be described using FIG. FIG. 11 is a diagram showing an example of a configuration of an estimation apparatus 100A according to a modification. As illustrated in FIG. 11, the estimation device 100A includes a communication unit 110, a storage unit 120A, and a control unit 130A. In addition, about the point similar to the estimation apparatus 100 which concerns on embodiment about the estimation apparatus 100A, description is abbreviate | omitted suitably.

(Storage unit 120A)
The storage unit 120A is implemented by, for example, a semiconductor memory device such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. As illustrated in FIG. 11, the storage unit 120A according to the modification includes a user information storage unit 121, an action information storage unit 122, a level estimation rule information storage unit 123A, and an anxiety action list information storage unit 124A.

(Level estimation rule information storage unit 123A)
The level estimation rule information storage unit 123A according to the modification stores various pieces of information regarding the level estimation rule (level estimation model). The level estimation rule information storage unit 123A stores various types of information used to estimate the level of the user's mental and physical state. For example, the level estimation rule information storage unit 123A stores various types of information used to estimate the level of each category. FIG. 12 is a diagram illustrating an example of a level estimation rule information storage unit according to a modification. For example, the level estimation rule information storage unit 123A stores an estimation model (estimation rule) based on a combination of “level” and “category”. More specifically, in the level estimation rule information storage unit 123A, information indicating an estimation rule based on the score of the user's mental and physical state level in the corresponding category is provided for each row in the row "level" and the column "category". (Rule information) is stored. In the example of FIG. 12, the level estimation rule information storage unit 123A has information indicating an estimation model of the anxiety level for each category based on the user's anxiety behavior in each box of row "level" and column "category". It is memorized.

  "Level" indicates the level of the user's mental and physical state in the corresponding category. In FIG. 12, “level” includes items such as “low (LV1)”, “middle (LV2)”, and “high (LV3)”. Note that the above is an example, and the “level” is not limited to three levels of low, medium, and high, and may be various stages (for example, ten stages).

  “Category” indicates a category for which the user's mental and physical state is to be estimated. In FIG. 12, “category” includes items such as “work (CTG1)”, “health (CTG2)”, “future (CTG3)”, and “unknown (CTG4)”. Note that the above is an example, and any category may be used as the "category" as long as the category is a target for estimating the physical and mental state of the user, such as "home" and "money".

  For example, in the box at which each level intersects with each category, a level estimation rule for estimating whether the user's mental and physical state exists at each level in the corresponding category is indicated. In the example of FIG. 12, the squares at which each level intersects with each category indicate level estimation rules for estimating whether the user's anxiety level falls under each level in the corresponding category.

  In the example of FIG. 12, in the box where the level "low (LV1)" and the category "job (CTG1)" intersect, "the number of times of anxiety behavior in a predetermined period is at least 1.2 times the normal time" is stored It is done. That is, in the example of FIG. 12, in the category “work (CTG1)”, the number of times of anxiety behaviors included in the action information of the target user in a predetermined period (for example, one week) is 1.2 times or more as normal. In this case, the target user's anxiety indicates "low (LV1)", that is, low level.

  For example, the estimation apparatus 100A calculates the number of anxiety behaviors included in the behavior information of the target user for the latest one week and the average number of one week of the number of anxiety behaviors included in the action information of the target user in the past one year (generally Time average). For example, the estimation apparatus 100A calculates the normal time average by dividing the number of anxiety behaviors included in the action information of the target user in the past one year by a predetermined period (“7” in the case of one week). Then, when the number of anxiety behaviors included in the action information of the target user for the latest one week is 1.2 times or more of the normal time average, the estimation apparatus 100A has low level of anxiety in the work of the target user. Estimate. For example, when the number of anxiety behaviors of the target user for the latest one week is 15 times and the average for the normal time is 10 times, the estimation device 100A determines that the latest one week is “1.5 (= 15 for normal time Calculated as "10)" times. In this case, since the last week is 1.2 times or more of the normal time, the estimation apparatus 100A estimates that the anxiety in the work of the target user is at a low level. The target on which the calculation of the normal hourly average is based is not limited to the past one year, but may be various periods such as the past six months or the past three years.

  In the example of FIG. 12, “the number of times of anxiety behavior in a predetermined period is twice or more as normal” is stored in the cell where the level “middle (LV2)” and the category “job (CTG1)” intersect There is. That is, in the example of FIG. 12, in the category “job (CTG1)”, when the number of times of anxious behavior included in the action information of the target user in a predetermined period (for example, one week etc.) The target user's anxiety indicates "medium (LV2)", ie, medium level. For example, in the case where the number of anxiety behaviors of the target user for the latest one week is 25 and the average for the normal time is 10, the estimation device 100A calculates “2.5 (= 25 for the normal one week)”. Calculated as "10)" times. In this case, since the last week is twice or more the normal time, the estimation apparatus 100A estimates that the anxiety in the work of the target user is at the middle level.

  In the example of FIG. 12, “the number of times of anxiety behavior for a predetermined period is five or more times the normal time” is stored in the cell where the level “high (LV3)” and the category “work (CTG1)” intersect There is. That is, in the example of FIG. 12, in the category “job (CTG1)”, when the number of anxiety behaviors included in the action information of the target user in a predetermined period (for example, one week) is five or more times the normal time, The target user's anxiety indicates "high (LV3)", that is, high level. For example, when the number of anxiety behaviors of the target user for the latest one week is 100 times and the average for the normal time is 10 times, the estimation device 100A determines that the latest one week is “10 (= 100/10) for the normal time. Calculated as “)”. In this case, the estimation apparatus 100A estimates that the anxiety in the work of the target user is at a high level because the last week is five times or more the normal time.

  As described above, the level estimation rule information storage unit 123A may store a model of level estimation based on relative comparison with that at normal time. The estimation device 100A may estimate the level of the user's mental and physical state based on the relative comparison with the normal time.

  Further, as shown in the column of the category "health (CTG2)", the level estimation rule information storage unit 123A may estimate the user's mental and physical state by using an absolute value such as the number of times of anxiety behavior. For example, “the number of times of anxiety behavior for a predetermined period of time is three or more” is stored in the box where the level “low (LV1)” and the category “health (CTG2)” intersect. In this case, when the number of anxiety behaviors included in the action information of the target user for the latest one week is three or more, the estimation device 100A estimates that the degree of anxiety in the work of the target user is low. As described above, the estimation device 100A may estimate the level of the physical and mental state of the user based on the comparison between the number of anxiety behaviors included in the behavior information of the target user and the predetermined threshold.

  The level estimation rule information storage unit 123A is not limited to the above, and may store various information according to the purpose. For example, the level estimation rule information storage unit 123A may store information when there is a specific action (specific action) for estimating that the user is uneasy only by the action. For example, when the target user is performing a specific action, the level estimation rule information storage unit 123A may store a rule for estimating the anxiety in a certain category of the target user as a high level. For example, when the action information of the target user includes an action using a specific query such as a query “retirement notice”, the level estimation rule information storage unit 123A presumes that the target user's anxiety at work is high level May be stored.

(Anxiety behavior list information storage unit 124A)
The anxiety action list information storage unit 124A according to the modification stores information on the anxiety behavior. For example, the anxiety action list information storage unit 124 </ b> A stores list (list) information on anxiety behaviors associated with each category. FIG. 13 is a diagram illustrating an example of the anxiety action list information storage unit according to the modification. In the example of FIG. 13, the anxiety action list information storage unit 124A includes information such as the action list set 124A-1, the action list set 124A-2, the action list set 124A-3, the action list set 124A-4, and the like for each category. (Table) is stored. The action list sets 124A-1 to 124A-4 and the like shown in FIG. 13 include items such as "category", "type", and "anxiety action list". In addition, the administrator of the estimation apparatus 100A may set the anxiety behavior of each category and type. Further, the estimation device 100A may use an anxiety action list common to all the categories.

  In the example of FIG. 13, in the action list set 124A-1, the anxiety action list corresponding to the category “work (CTG1)” is stored. In addition, in the action list set 124A-2, the anxiety action list corresponding to the category "health (CTG2)" is stored. In addition, in the action list set 124A-3, the anxiety action list corresponding to the category “future (CTG3)” is stored. In addition, in the action list set 124A-4, the anxiety action list corresponding to the category "unknown (CTG 4)" is stored.

  "Type" indicates the type of action for acquiring (counting) anxiety behaviors. In the example of FIG. 13, “type” is illustrated as “search”, “purchase”, “browse”, “insurance participation”, “position” but various types such as “post” may be included. . Further, the “type” may include weather conditions and the like when the user performs an action, such as “weather” and “air temperature”. That is, any kind of information may be included in the anxiety behavior as long as the information is related to the behavior of the user.

  In the "anxiety behavior list", a list of the corresponding anxiety behaviors is stored. The “anxiety behavior list” includes items such as “anxiety behavior # 1”, “anxiety behavior # 2”, and “anxiety behavior # 3”. Although FIG. 13 illustrates the anxiety behaviors # 1 to # 3 etc., when there are four or more anxiety behaviors corresponding to the category and type, four anxiety behaviors # 4 and five etc. The above anxiety behavior may be stored.

  For example, in the example shown in FIG. 13, in the category “job (CTG1)”, the anxiety action corresponding to the type “search” is a search using a query such as “query UQ1”, “query UQ2” or “query UQ3” To indicate that In FIG. 13, although it is illustrated by an abstract code such as “query UQ1”, “query UQ1” or the like may be a query (character information) related to work such as “performance slump” or “recession”. The “query UQ1” or the like is not limited to the text information, and may be information such as an image.

  For example, in the example shown in FIG. 13, in the category “job (CTG1)”, the anxiety behavior corresponding to the type “purchase” is a transaction target such as “trading target UT1”, “trading target UT2” or “trading target UT3” Indicates that it is a purchase of For example, trading objects may be various trading objects such as goods or services. In FIG. 13, although it is illustrated by an abstract code such as “Trading target UT1”, “Trading target UT1” and the like are “P drug” which is a sleep introducing agent and “F book” which is a fortune telling book It may be a transaction object.

  For example, in the example shown in FIG. 13, in the category “job (CTG1)”, the anxiety action corresponding to the type “view” is viewing of content such as “content UC1”, “content UC2” or “content UC3” Indicates that. In FIG. 13, although it is illustrated by an abstract code such as “content UC1”, “content UC1” and the like are contents such as “N news” which is news of a large scale disaster and “F site” of a fortune telling site Web page). Note that “content UC1” and the like are not limited to web pages and the like, and may be various contents such as moving pictures and sounds.

  For example, in the example shown in FIG. 13, in the category “job (CTG1)”, anxiety behavior corresponding to the type “insurance participation” is participation in insurance such as “insurance IS1”, “insurance IS2” or “insurance IS3” To indicate that In FIG. 13, although illustrated with an abstract code such as “insurance IS1”, “insurance IS1” or the like may be an insurance such as “XX life insurance” or “YY cancer insurance”.

  For example, in the example shown in FIG. 13, in the category “work (CTG1)”, the anxiety action corresponding to the type “position” is the user's move to a position such as “position UL1”, “position UL2” or “position UL3”. Indicates that it is a visit. In FIG. 13, although it is illustrated by an abstract code such as “location UL1”, “location UL1” and the like indicate specific latitude and longitude, an address, etc. of the place where the job security office is located or the place where the hospital is located. It may be information of

  The anxiety action list information storage unit 124 </ b> A is not limited to the above, and may store various information according to the purpose. For example, the anxiety action list information storage unit 124A may store the anxiety action list associated with each user attribute. For example, the anxiety action list information storage unit 124A may store an anxiety action list associated with each user attribute such as a 20's male, a 20's female, and a 30's male.

(Control unit 130A)
The control unit 130A is a controller, and for example, various programs (corresponding to an example of an estimation program and a generation program) stored in a storage device in the estimation device 100A by a CPU, an MPU, etc. It is realized by being executed as The control unit 130A is a controller, and is realized by, for example, an integrated circuit such as an ASIC or an FPGA.

  As illustrated in FIG. 11, the control unit 130A includes an acquisition unit 131, an estimation unit 133A, a determination unit 135A, a provision unit 136A, and a generation unit 136, and has functions and operations of estimation processing described below. Achieve or execute. Control unit 130A is not limited to the configuration shown in FIG. 11, and may be another configuration as long as it performs a estimation process to be described later.

(Acquisition unit 131)
The acquisition unit 131 according to the modification performs the same process as the acquisition unit 131 according to the embodiment. The acquisition unit 131 acquires various pieces of information as the acquisition unit 131 according to the embodiment.

  In the example of FIG. 10, the acquisition unit 131 acquires the behavior information of the user U2 from the terminal device 10 used by the user U2. The acquisition unit 131 acquires the action information of the user U2 in a predetermined period (for example, one week). For example, the estimation device 100A acquires the action information ADT21 and ADT22 of the user U2 of the latest one week. The acquisition unit 131 acquires, from the terminal device 10, behavior information of the user U2 on the Internet. The acquisition unit 131 acquires the action information ADT 21 indicating that the user U 2 browsed the content # 21. The acquisition unit 131 acquires behavior information ADT22 indicating that the user U2 has purchased (purchased) the product # 22.

(Estimation unit 133A)
The estimation unit 133A according to the modification performs the same process as the estimation unit 133 according to the embodiment. The estimation unit 133A performs various estimations in the same manner as the estimation unit 133 according to the embodiment. The estimation unit 133A estimates various information as in the estimation unit 133 according to the embodiment.

  In the example of FIG. 10, the estimation unit 133A estimates the anxiety level of the user U2 for each category. The estimation unit 133A estimates the anxiety level of the user U2 using the information stored in the level estimation rule information storage unit 123A and the anxiety action list information storage unit 124A. The estimation unit 133A estimates the anxiety level of the user U2 in each category using the counted number of times of the anxiety activity of each category and the information of the level estimation rule information storage unit 123A.

  In the example of FIG. 10, the estimation unit 133A calculates the number of anxiety behaviors (the number of anxiety) included in the action information of the target user for the last week, and the threshold of each level (three times, five times, ten times, etc.) Compare For example, when the number of times of anxiety is equal to or higher than the high level threshold “10”, the estimation unit 133A estimates that the anxiety in the health of the target user is high. Also, for example, when the number of anxiety is less than the high level threshold “10” and equal to or higher than the middle level threshold “5”, the estimation unit 133A estimates that the health user's anxiety is at the middle level. Also, for example, when the number of anxiety is less than the middle level threshold “5” and the low level threshold “3”, the estimation unit 133A estimates that the health user's anxiety is low.

  In the example of FIG. 10, the estimation unit 133A determines that the number of times of anxiety behavior corresponding to the category "health" is "12" times and the "predetermined period of time" of the level "high (LV3)" of the category "health (CTG2)". It is determined that the condition of “the number of anxiety behaviors is 10 or more” is satisfied. Therefore, as shown in the estimated level list UL21, the estimation unit 133A estimates that the anxiety level of health of the user U2 is high.

  Further, in the example of FIG. 10, since the estimation unit 133A determines that the number of anxiety behaviors corresponding to the categories “work”, “future”, and “unknown” is “0” times, the work of the user U2, the future, and the unknown It is estimated that there is no anxiety. Therefore, as shown in the estimated level list UL21, the estimation unit 133A estimates the work, future, and unknown anxiety levels of the user U2 as "none (LV0)."

(Determination unit 135A)
The determination unit 135A according to the modification performs the same process as the determination unit 135 according to the embodiment. The determination unit 135A performs various determinations in the same manner as the determination unit 135 according to the embodiment. The determination unit 135A determines various pieces of information in the same manner as the determination unit 135 according to the embodiment.

  In the example of FIG. 10, the determination unit 135A determines the premium of the user U2 in accordance with the anxiety level of the user U2 estimated by the estimation unit 133A. As shown in the user information storage unit 121, the determination unit 135A recommends that the user U2 join according to the anxiety level of the user U2 estimated by the estimation unit 133A because the user U2 has not subscribed to insurance. Determine the premium to be presented to

  For example, since the determination unit 135A estimates that the anxiety level of the user U2 is high by the estimation unit 133A, the determination unit 135A determines to present the user U2 a premium higher than the premium paid by the other users. For example, the determination unit 135A determines that the user U2 is to be provided with information to recommend subscription to the insurance to which the user having the user attribute similar to that of the user U2 has subscribed. In addition, since the determination unit 135A estimates that the anxiety level of health of the user U2 is high by the estimation unit 133A, the determination unit 135A determines to present the user U2 a premium higher than the premium paid by the user whose user attribute is similar to that of the user U2. Do.

  In the example of FIG. 10, the determination unit 135A determines to present the C cancer insurance to which the user U5 who is a twentieth woman is subscribed to the user U2 with a premium higher than the premium paid by the user U5. For example, the determination unit 135A determines to recommend the C cancer insurance to the user U2 as a premium UZ yen higher than the premium Z yen paid by the user U5.

(Providing portion 136A)
The providing unit 136A according to the modification performs the same process as the providing unit 136 according to the embodiment. The providing unit 136A performs various offers in the same manner as the providing unit 136 according to the embodiment. The providing unit 136A provides various information as the providing unit 136 according to the embodiment.

  In the example of FIG. 10, the providing unit 136A provides information to the terminal device 10 used by the user U2. The providing unit 136A transmits, to the terminal device 10 used by the user U2, information for inviting the participation in C cancer insurance. The providing unit 136A distributes an advertisement (advertisement ACX) recommending C cancer insurance to the terminal device 10 used by the user U2 as a premium UZ yen higher than the premium Z yen paid by the user U5. The providing unit 136A distributes the advertisement ACX including the information indicating the C cancer insurance and the information indicating that the premium is UZ yen to the terminal device 10 used by the user U2.

(Generation unit 136)
The generation unit 136 performs various generations. For example, the generation unit 136 generates a model. For example, the generation unit 136 generates various models as shown in the model information storage unit 123 in FIG.

  For example, the generation unit 136 generates a model as shown in the model information storage unit 122 in FIG. 6 using learning data collected by the estimation process as shown in FIG. For example, the generation unit 136 may generate a model corresponding to each user or each category, with the anxiety level estimated in FIG. 10 as the correct answer information. For example, the generation unit 136 may generate a model using, as learning data, a combination of the action information of the target user and the estimated anxiety level shown in FIG.

  For example, the generation unit 136 may use the anxiety level of each category as shown in the estimated level list UL21 in FIG. 10 as correct answer information. For example, the generation unit 136 learns the anxiety level of each category corresponding to the input data and the behavior information ATD21, ADT22, etc. of the user U2 in a predetermined period shown in FIG. As, the model corresponding to each user or each category may be generated. For example, in the case of the input data of the user U2, the generation unit 136 is a model that outputs a high score when the input data of the user U2 is input because the anxiety level of the category "health" is estimated to be high level. M2-2 (see FIG. 6) may be generated.

  As described above, the generation unit 136 may generate a model corresponding to the user U2 that outputs a high score (for example, “1”) when the input data of the user U2 is input, and the category “health”. For example, when the input data of the user U2 is input to the model M2-2, the generation unit 136 performs learning processing so that the score output by the model M2-2 approaches “1”.

  Further, for example, in the case of the input data of the user U2, the generation unit 136 estimates that the anxiety level of the category “work” is none “LV0”, and therefore the score is low when the input data of the user U2 is input. May be generated (see FIG. 6). As described above, the generation unit 136 may generate a model corresponding to the user U2 that outputs a low score (for example, “0”) when the input data of the user U2 is input, and the category “task”. For example, when the input data of the user U2 is input to the model M2-1, the generation unit 136 performs learning processing so that the score output from the model M2-1 approaches “0”.

  Also, for example, when the anxiety level is estimated to be low level or middle level, the generation unit 136 generates predetermined level information (see the level information storage unit 124 in FIG. 7) when the behavior information is input. The model may be generated to output the corresponding score. For example, when the anxiety level of the category "work" is estimated to be low, the generation unit 136 determines that "0.2 or more 0. 0." corresponding to the low level of the category "work" when the behavior information is input. A model corresponding to the category “task” that outputs a score of “less than 5” (eg, “0.4” or the like) may be generated. For example, when behavior information in which the anxiety level of the category “work” is estimated to be low is input to the model, the generation unit 136 learns so that the score output by the model approaches “0.4”. Do the processing.

  Also, for example, when the anxiety level of the category “work” is estimated to be the middle level, the generation unit 136 determines “0.5 or more corresponding to the middle level of the category“ work ”when the behavior information is input. A model corresponding to the category "task" which outputs a score (e.g., "0.7" or the like) or the like less than 0.8 may be generated. For example, when behavior information in which the anxiety level of the category “work” is estimated to be middle level is input to the model, the generation unit 136 learns so that the score output by the model approaches “0.7”. Do the processing.

  For example, the generation unit 136 generates the models M1-1 to M3-4 and the like as illustrated in FIG. 6, and stores the generated models M1-1 to M3-4 and the like in the storage unit 120A. The generation unit 136 may generate the models M1-1 to M3-4 and the like using any learning algorithm. For example, the generation unit 136 generates the models M1-1 to M3-4 and the like using a learning algorithm such as a neural network, a support vector machine (SVM), clustering, and reinforcement learning. As an example, when the generation unit 136 generates the models M1-1 to M3-4 and the like using a neural network, the models M1-1 to M3-4 and the like have an input layer including one or more neurons and one or more input layers. It has an intermediate layer containing neurons and an output layer containing one or more neurons.

  The generation unit 136 generates a model, and stores the generated model in the storage unit 120A. Specifically, the generation unit 136 is an input layer to which the user's action information is input, an output layer, and a first element that is any layer from the input layer to the output layer and belongs to a layer other than the output layer. , And a second element whose value is calculated based on the first element and the weight of the first element, and each element belonging to each layer other than the output layer is By performing an operation based on the first element and the weight of the first element as one element, a model is generated in which the value of the score used to estimate the user's mental and physical state is output from the output layer.

  In addition, about the learning method of a model, it is not limited to the method mentioned above, Arbitrary well-known techniques are applicable. Note that generation of each model may be performed using various conventional techniques related to machine learning as appropriate. For example, model generation may be performed using techniques related to machine learning of supervised learning such as support vector machines. Also, for example, model generation may be performed using a technique related to machine learning of unsupervised learning. For example, model generation may be performed using deep learning techniques. For example, generation of a model may be performed using various deep learning techniques such as DNN, Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), etc. as appropriate. In addition, the description regarding the production | generation of the said model is an illustration, and the production | generation of a model may be performed by the learning method suitably selected according to the information etc. which can be acquired. That is, when generation unit 136 is capable of learning models M1-1 to M3-4 and the like so as to output a score corresponding to correct answer information when action information included in learning data is input, The models M1-1 to M3-4 and the like may be generated by any appropriate method.

  Thus, the generation unit 136 generates a model based on learning data in which the action information of the user and the anxiety level of each category are correct information. For example, the generation unit 136 generates a model using a feature amount related to behavior information. For example, the generation unit 136 generates a model using the feature amount related to the content of the user's action. For example, the generation unit 136 generates a model using a feature amount related to a user's action target such as a query used by the user or a transaction target purchased by the user. Also, for example, the generation unit 136 generates a model using feature information related to time information, position information, and weather information in the action of the user. For example, the generation unit 136 learns a feature related to the action included in the action information of the user collected in the example illustrated in FIG. 10, and generates a model for estimating the mental state of the user who is performing the action. The feature included in the action information learned by the generation unit 136 may be input to the generation unit 136 by a person such as a manager of the estimation device 100A, or the estimation device 100A is automatically learned (extracted) It is also good.

  Also, when the estimation device 100A and the estimation device 100 are separate bodies, the estimation device 100A may provide model information to the estimation device 100. In this case, for example, the estimation device 100A may be the information provision source device 20 in the estimation system 1 according to the embodiment.

[5. About level estimation rule information]
Further, the estimation apparatus 100A may use level estimation rule information associated with each user attribute. For example, the estimation device 100A may use level estimation rule information associated with each user attribute, such as a 20's male, a 20's female, a 30's male and so on. This point will be described with reference to FIG. FIG. 14 is a diagram illustrating another example of the level estimation rule information storage unit according to the modification.

(Level estimation rule information storage unit 123B)
The level estimation rule information storage unit 123B according to the modification stores various pieces of information related to the level estimation rule. The level estimation rule information storage unit 123B stores various information used to estimate the level of the user's mental and physical state. For example, the level estimation rule information storage unit 123B stores various types of information used to estimate the level of each category. FIG. 14 is a diagram illustrating an example of a level estimation rule information storage unit according to a modification. In the example of FIG. 14, the level estimation rule information storage unit 123B includes information (table) for each user attribute, such as the level estimation rule set 123B-1, the level estimation rule set 123B-2, and the level estimation rule set 123B-3. Remember. The level estimation rule sets 123B-1 to 123B-3 and the like shown in FIG. 14 have items of "user attribute", and an estimation model (estimation rule) by combination of "level" and "category" is stored. Specifically, in the level estimation rule set 123B-1 to 123B-3, etc., each cell in the row "level" and the column "category" is based on the score of the user's mental and physical state level in the corresponding category. Information (rule information) indicating an estimation rule is stored. In the example of FIG. 14, in the level estimation rule information storage unit 123B, information indicating an estimation model of the anxiety level for each category based on the anxiety behavior of the user is included in each row that is a row “level” and a column “category”. It is memorized.

  In the example of FIG. 14, the level estimation rule set 123B-1 stores the anxiety action list corresponding to the user attribute “male in twenties”. Further, the level estimation rule set 123B-2 stores the anxiety action list whose user attribute corresponds to "20's female". In the level estimation rule set 123B-3, the anxiety action list corresponding to the user attribute "male in thirties" is stored.

  "Level" indicates the level of the user's mental and physical state in the corresponding category. In FIG. 14, “level” includes items such as “low (LV1)”, “middle (LV2)”, and “high (LV3)”. Note that the above is an example, and the “level” is not limited to three levels of low, medium, and high, and may be various stages (for example, ten stages).

  “Category” indicates a category for which the user's mental and physical state is to be estimated. In FIG. 14, the “category” includes items such as “work (CTG1)”, “health (CTG2)”, “future (CTG3)”, and “unknown (CTG4)”. Note that the above is an example, and any category may be used as the "category" as long as the category is a target for estimating the physical and mental state of the user, such as "home" and "money".

  For example, in the box at which each level intersects with each category, a level estimation rule for estimating whether the user's mental and physical state exists at each level in the corresponding category is indicated. In the example of FIG. 14, level squares at which each level intersects with each category indicate level estimation rules for estimating whether the user's anxiety level falls under each level in the corresponding category.

  When using the information of the level estimation rule information storage unit 123B of FIG. 14, the estimation apparatus 100A estimates the mental and physical state of the target user using the level estimation rule set corresponding to the user attribute of the target user. For example, when the user attribute of the target user is “20's female”, the estimation apparatus 100A estimates the mental and physical state of the target user using the level estimation rule stored in the level estimation rule set 123B-2. Thus, the estimation apparatus 100A can appropriately estimate the mental and physical state of the target user according to the user attribute to which the target user belongs.

  Further, for example, the estimation apparatus 100A may generate a model for each user attribute, with the mind-body state estimated using the level estimation rule for each user attribute as correct answer information. By using such a model for each user attribute, the estimation apparatus 100A can estimate the mental and physical state of the target user based on comparison with the user group in the user attribute to which the target user belongs.

[6. About combination of embodiment and modification]
For example, when performing estimation using the generated model, the estimation device 100A may perform the same process as the estimation device 100 according to the embodiment. Also, in this case, the storage unit 120A of the estimation device 100A may include the model information storage unit 123, the level information storage unit 124, and the content information storage unit 125. Further, the control unit 130A of the estimation device 100A may include the selection unit 132. For example, when the estimation device 100A performs estimation using a model, the estimation device 100A may be integral with the estimation device 100 according to the embodiment.

  In this case, for example, the estimation apparatus 100A performs estimation processing using information of the level estimation rule information storage unit 123A or the anxiety action list information storage unit 124A for a target user or target category before the model is generated. For example, when the user U1 is before model generation, the estimation device 100A estimates the mental and physical state in the category "work" of the user U1 using the information of the level estimation rule information storage unit 123A and the anxiety action list information storage unit 124A. . For example, as shown in FIG. 10, when the user U2 is before model generation, the estimation apparatus 100A uses the information of the level estimation rule information storage unit 123A and the anxiety action list information storage unit 124A of the user U2's mental and physical state. presume.

  Further, the estimation apparatus 100A performs estimation processing using the information of the model information storage unit 123 or the level information storage unit 124 for the target user or target category after the model is generated. For example, as shown in FIG. 1, when the user U1 has generated a model, the estimation device 100A estimates the mental and physical state of the user U1 using the information of the model information storage unit 123 and the level information storage unit 124. For example, as illustrated in FIG. 1, the estimation device 100A estimates whether the mind-body state of the category “work” of the user U1 is anxiety using the model M1-1 corresponding to the user U1. For example, when the user U2 has generated a model, the estimation device 100A estimates the mental and physical state of the user U2 using the information of the model information storage unit 123 and the level information storage unit 124. For example, when the user U2 is after generation of a model, the estimation device 100A uses the model M2-1 corresponding to the user U2 to estimate whether the mind-body state in the category "work" of the user U2 is anxiety.

  For example, the estimation apparatus 100A may generate models in order from a user or category in which learning data for generating a model is equal to or greater than a predetermined threshold. In this case, the estimation apparatus 100A performs estimation using a model for an estimation target for which a model has been generated, and for the estimation target for which a model is not generated, the estimation apparatus 100A of the level estimation rule information storage unit 124A or the anxiety action list information storage unit 124A The information may be used for estimation.

[7. effect〕
As described above, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification include the acquisition unit 131 and the estimation units 133 and 133A. The acquisition unit 131 acquires the action information of the target user. In addition, the estimation units 133 and 133A estimate the mental and physical state of the target user based on the action information indicating the action of the target user acquired by the acquisition unit 131.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user based on the action of the target user, and can appropriately estimate the state of the user. it can.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires behavior information of the target user on the Internet.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the physical and mental state of the target user based on the action of the target user on the Internet, and appropriately estimate the state of the user can do.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires behavior information including query information used for a search by the target user.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user based on the query information used for the search by the target user, and the user's state is appropriate. Can be estimated.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires behavior information including purchase information of a target user.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user based on the purchase information of the target user, and appropriately estimate the state of the user. Can.

  In addition, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires behavior information including position information of a target user.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user based on the position information of the target user, and appropriately estimate the state of the user. Can.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires behavior information including weather information at a position corresponding to a target user.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user based on the weather information at the position corresponding to the target user, and the user's state is appropriate Can be estimated.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the estimation units 133 and 133A estimate whether the physical and mental state of the target user is another state different from the normal state.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate whether the physical and mental state of the target user is another state different from the normal state, and the user's state is appropriately determined. It can be estimated.

  Further, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the estimation units 133 and 133A estimate whether the physical and mental state of the target user is a negative state different from the normal state.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate whether the physical and mental state of the target user is a negative state different from the normal state, and the user's state is appropriately determined. It can be estimated.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the estimation units 133 and 133A estimate whether or not the physical and mental state of the target user is anxiety.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate whether the mind and body state of the target user is anxious and can appropriately estimate the state of the user.

  Further, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the acquisition unit 131 acquires a model used for estimation of the user's physical and mental state from the user's action information. The estimation units 133 and 133A estimate the mental and physical state of the target user based on the target user's behavior information and the model.

  Thus, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can estimate the mental and physical state of the target user using the model, and can appropriately estimate the state of the user.

  Further, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the acquisition unit 131 uses a model used to estimate which one of the multiple levels of the mind and body state is the user's mind and body state. get. The estimation units 133 and 133A estimate the level of the physical and mental state of the target user based on the score output by the model when the target user's action information is input.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification estimate the level of the mental and physical state of the target user based on the score output by the model when the action information of the target user is input. And the user's state can be properly estimated.

  Further, the estimation device 100 according to the embodiment and the estimation device 100A according to the modification include a selection unit 132. The selection unit 132 selects one model based on a predetermined condition among a plurality of models generated for each predetermined object. The estimation units 133 and 133A estimate the mental and physical state of the target user based on the one model selected by the selection unit 132.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification estimate the level of the mental and physical state of the target user based on the score output by the model when the action information of the target user is input. And the user's state can be properly estimated.

  Further, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the acquisition unit 131 acquires a plurality of models generated for each user. The selection unit 132 selects one model corresponding to the target user among the plurality of models generated for each predetermined target. The estimation units 133 and 133A estimate the mental and physical state of the target user based on the one model selected by the selection unit 132.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification select one model corresponding to the target user among the plurality of models generated for each predetermined target, and select one Based on the model, it is possible to estimate the level of the mental and physical state of the target user regarding a predetermined target corresponding to one model, and it is possible to appropriately estimate the state of the user.

  In addition, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the acquisition unit 131 acquires a plurality of models generated for each of a plurality of categories to be subjected to estimation of the user's mental and physical state. The selection unit 132 selects one of the plurality of models that corresponds to a target category to be estimated. The estimation units 133 and 133A estimate the mental and physical state of the target user in the target category based on the one model selected by the selection unit 132.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification select one model corresponding to the target user among the plurality of models generated for each of the plurality of categories, and select one Based on the model, it is possible to estimate the level of the target user's mental and physical state related to the category corresponding to one model, and to appropriately estimate the user's state.

  Further, the estimation device 100 according to the embodiment and the estimation device 100A according to the modification include a selection unit 134. The selection unit 134 selects the processing target user who is the user who is the target of the predetermined processing based on the information indicating the user's health condition and the mental and physical condition of the user estimated by the estimation units 133 and 133A. .

  Thereby, the estimation device 100 according to the embodiment and the estimation device 100A according to the modification are predetermined based on the information indicating the user's health condition and the physical and mental condition of the user estimated by the estimation units 133 and 133A. By selecting the processing target user who is the user who is the target of processing, it is possible to appropriately select the processing target user according to the user's health and the state of mind and body.

  In addition, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the acquisition unit 131 acquires information indicating the health condition of the target user. The selection unit 134 selects the target user as the processing target user when the difference between the health state of the target user and the physical and mental state of the target user satisfies a predetermined condition.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification set the target user as the processing target user when the difference between the health state of the target user and the physical and mental state of the target user satisfies a predetermined condition. By selecting, it is possible to appropriately select the user to be processed according to the user's health and both mental and physical states.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the selection unit 134 is in the normal state of the target user's health and the other in which the target user's physical and mental state differs from the normal state. If it is in the state, the target user is selected as the processing target user.

  Accordingly, when the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification are in the normal state of the target user's health and in the other state different from the normal state of the target user's physical and mental state By selecting the target user as the processing target user, it is possible to appropriately select the user to be the processing target according to both the user's health and the state of mind and body.

  Further, the estimation device 100 according to the embodiment and the estimation device 100A according to the modification include determination units 135 and 135A. The determination units 135 and 135A determine the premium to be applied to the target user based on the physical and mental state of the target user estimated by the estimation units 133 and 133A.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can determine the premium to be applied to the target user based on the estimated physical and mental state of the target user, and is appropriate according to the state of the user Can provide various insurance services.

  The estimation device 100 according to the embodiment and the estimation device 100A according to the modification include providing units 136 and 136A. The providing units 136 and 136A provide a service based on the physical and mental state of the target user estimated by the estimating units 133 and 133A.

  Thereby, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can provide a service to the target user based on the estimated physical and mental state of the target user, and an appropriate service can be provided according to the state of the user. It can be made available.

  Further, in the estimation device 100 according to the embodiment and the estimation device 100A according to the modification, the providing units 136 and 136A provide an information providing service based on the physical and mental state of the target user estimated by the estimation units 133 and 133A.

  As a result, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can provide appropriate information to the target user based on the estimated physical and mental state of the target user, depending on the state of the user. Appropriate information provision service can be provided.

  Further, in the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification, the providing units 136 and 136A provide a content distribution service based on the physical and mental state of the target user estimated by the estimation units 133 and 133A.

  As a result, the estimation apparatus 100 according to the embodiment and the estimation apparatus 100A according to the modification can deliver appropriate content to the target user based on the estimated physical and mental state of the target user, according to the state of the user To provide appropriate content distribution services.

[8. Hardware configuration]
The estimation device 100 according to the embodiment described above and the estimation device 100A according to the modification are realized by, for example, a computer 1000 configured as shown in FIG. FIG. 15 is a hardware configuration diagram showing an example of a computer that implements the function of the estimation device. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM 1300, an HDD 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

  The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 to control each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 starts up, a program depending on the hardware of the computer 1000, and the like.

  The HDD 1400 stores a program executed by the CPU 1100, data used by the program, and the like. The communication interface 1500 receives data from another device via the network N, sends the data to the CPU 1100, and transmits data generated by the CPU 1100 to the other device via the network N.

  The CPU 1100 controls an output device such as a display or a printer and an input device such as a keyboard or a mouse via the input / output interface 1600. The CPU 1100 acquires data from an input device via the input / output interface 1600. The CPU 1100 also outputs the generated data to the output device via the input / output interface 1600.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the CPU 1100 with the program via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a digital versatile disc (DVD) or a phase change rewritable disc (PD), a magneto-optical recording medium such as a magneto-optical disk (MO), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  For example, when the computer 1000 functions as the estimation device 100 according to the embodiment or the estimation device 100A according to the modification, the CPU 1100 of the computer 1000 reads a program (for example, model M1-1 (model data MDT1- By performing 1), the functions of the control units 130 and 130A are realized. The CPU 1100 of the computer 1000 reads and executes these programs (for example, model M1-1 (model data MDT1-1)) from the recording medium 1800, but as another example, these may be transmitted from another device via the network N. You may get a program of

  Although some of the embodiments and modifications of the present application have been described above in detail based on the drawings, these are merely examples, and various aspects can be obtained based on the knowledge of those skilled in the art, including the aspects described in the rows of the disclosure. It is possible to carry out the present invention in other forms that have been modified and improved.

[9. Other]
Further, among the processes described in the embodiment and the modification, all or part of the process described as being automatically performed can be manually performed, or it is described as manually performed. All or part of the treatment may be performed automatically by known methods. In addition, information including processing procedures, specific names, various data and parameters shown in the above-mentioned documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

  Further, each component of each device illustrated is functionally conceptual, and does not necessarily have to be physically configured as illustrated. That is, the specific form of the distribution and integration of each device is not limited to that shown in the drawings, and all or a part thereof may be functionally or physically dispersed in any unit depending on various loads, usage conditions, etc. It can be integrated and configured.

  Moreover, it is possible to combine suitably the embodiment and modification which were mentioned above in the range which does not contradict process content.

  In addition, the "section (module, unit)" described above can be read as "means" or "circuit". For example, the acquisition unit can be read as an acquisition unit or an acquisition circuit.

DESCRIPTION OF SYMBOLS 1 estimation system 100 estimation device 121 user information storage unit 122 behavior information storage unit 123 model information storage unit 124 level information storage unit 125 content information storage unit 130 control unit 131 acquisition unit 132 selection unit 133 estimation unit 134 selection unit 135 determination unit 136 Provider 10 Terminal device 50 Information provider device N Network

Claims (23)

An acquisition unit for acquiring action information of a target user;
An estimation unit configured to estimate the mental and physical state of the target user based on the action information indicating the action of the target user acquired by the acquisition unit;
An estimation apparatus comprising: The acquisition unit
The estimation device according to claim 1, wherein the behavior information of the target user on the Internet is acquired. The acquisition unit
The estimation apparatus according to claim 2, wherein the action information including the query information used for the search by the target user is acquired. The acquisition unit
The estimation apparatus according to any one of claims 1 to 3, wherein the action information including purchase information of the target user is acquired. The acquisition unit
The estimation apparatus according to any one of claims 1 to 4, wherein the action information including position information of the target user is acquired. The acquisition unit
The estimation apparatus according to any one of claims 1 to 5, wherein the behavior information including weather information at a position corresponding to the target user is acquired. The estimation unit
The estimation apparatus according to any one of claims 1 to 6, wherein it is estimated whether the physical and mental state of the target user is another state different from a normal state. The estimation unit
The estimation apparatus according to claim 7, wherein it is estimated whether the mental and physical state of the target user is a negative state different from a normal state. The estimation unit
The estimation apparatus according to claim 8, wherein it is estimated whether the mind-body state of the target user is anxiety. The acquisition unit
Obtain a model used to estimate the user's mental and physical state from the user's action information,
The estimation unit
The estimation apparatus according to any one of claims 1 to 9, wherein the mental and physical state of the target user is estimated based on the action information of the target user and the model. The acquisition unit
Acquiring, from among the plurality of levels of the mind-body state, the model used to estimate which level the mind-body state of the user is.
The estimation unit
The estimation apparatus according to claim 10, wherein the level of the mind-body state of the target user is estimated based on a score output by the model when the action information of the target user is input. A selection unit that selects one model based on a predetermined condition among a plurality of models generated for each predetermined object,
And further
The estimation unit
The estimation apparatus according to claim 10, wherein the mental and physical state of the target user is estimated based on the one model selected by the selection unit. The acquisition unit
Acquiring the plurality of models generated for each of the users,
The selection unit is
The one model corresponding to the target user is selected from the plurality of models,
The estimation unit
The estimation apparatus according to claim 12, wherein the mental and physical state of the target user is estimated based on the one model selected by the selection unit. The acquisition unit
Acquiring the plurality of models generated for each of a plurality of categories for which the user's mental and physical state is to be estimated;
The selection unit is
Among the plurality of models, the one model corresponding to a target category to be estimated is selected;
The estimation unit
The estimation apparatus according to claim 12 or 13, wherein the mental and physical state of the target user in the target category is estimated based on the one model selected by the selection unit. A selection unit that selects a processing target user who is a user who is a target for performing a predetermined process based on information indicating a user's health condition and the mind and body condition of the user estimated by the estimation unit;
The estimation device according to any one of claims 1 to 14, further comprising: The acquisition unit
Acquiring information indicating the health condition of the target user;
The selection unit is
The estimation device according to claim 15, wherein the target user is selected as the processing target user when a difference between the health state of the target user and the physical and mental state of the target user satisfies a predetermined condition. The selection unit is
The target user is selected as the processing target user when the health state of the target user is a normal state and the physical and mental state of the target user is another state different from the normal state. The estimation device according to claim 16. A determination unit that determines a premium to be applied to the target user based on the physical and mental state of the target user estimated by the estimation unit;
The estimation device according to any one of claims 1 to 17, further comprising: A providing unit that provides a service based on the mind-body state of the target user estimated by the estimation unit;
The estimation device according to any one of claims 1 to 18, further comprising: The providing unit is
20. The estimation apparatus according to claim 19, providing an information providing service based on the mind-body state of the target user estimated by the estimation unit. The providing unit is
The estimation device according to claim 19 or 20, which provides a content distribution service based on the mind-body state of the target user estimated by the estimation unit. A computer implemented estimation method,
An acquisition step of acquiring action information of a target user;
An estimation step of estimating the mental and physical state of the target user based on the action information indicating the action of the target user acquired by the acquisition step;
Estimation method characterized by including. An acquisition procedure for acquiring action information of a target user;
An estimation procedure for estimating the mental and physical state of the target user based on the action information indicating the action of the target user acquired by the acquisition procedure;
An estimation program that causes a computer to execute.

Images