JP7586846B2 - Social participation behavior analysis device and method - Google Patents

Description

The present invention relates to technology for analyzing users' social participation behavior.

In order to evaluate a person's (user's) activity level, a technology has been developed that measures the person's activity level using a three-dimensional acceleration sensor or the like, corrects the measured activity level with environmental information such as temperature, and evaluates based on the corrected activity level. One example of such an activity level evaluation system is described in JP 2007-275260 A (Patent Document 1). Patent Document 1 describes how "a corrected measurement value is obtained based on activity intensity data and activity time period data, and correction is made with at least external environmental information and fixed information, and the corrected measurement value is evaluated using an evaluation table described below, and the evaluation result is formed into specified display screen data and provided to the display device 52" (paragraph 0017).

JP 2007-275260 A

In Patent Document 1, activity level is corrected using an artificially set regression model based on known knowledge that temperature and activity level are closely related. Therefore, appropriate correction cannot be made for activities or social participation behaviors for which knowledge is unknown. If an attempt is made to correct social participation information whose relationship with environmental information such as travel time or travel distance is unknown, it is necessary to perform correction after analyzing the relationship between each piece of social participation information and environmental information based on data. Therefore, an object of the present invention is to calculate social participation information more accurately. Note that this calculation includes correction.

One representative aspect of the present invention to solve the above problem is to correct the user's social participation information based on a regression model for each type of social participation that corresponds to the environmental information.

A more specific aspect of the present invention is a social participation behavior analysis device that analyzes a user's social participation behavior, the social participation behavior analysis device having an input unit that receives sensor information indicating the behavior of multiple users and environmental information indicating the environment at the time of the behavior, a social participation information identification unit that identifies social participation information indicating the user's social participation behavior from the sensor information, a model generation unit that generates a regression model of the social participation information against the environmental information for each type of social participation behavior, a correction amount calculation unit that calculates a correction amount of the social participation information corresponding to the environmental information based on the regression model, and a social participation information correction unit that corrects the social participation information using the correction amount.

The present invention also includes a social participation behavior analysis method using a social participation behavior analysis device, a program for causing the social participation behavior analysis device to function as a computer, and a recording medium for storing this program. Furthermore, a social participation analysis system including the social participation behavior analysis device and a user terminal is also an aspect of the present invention. Furthermore, the present invention also includes a program (app) for causing the user terminal to function as a computer.

The present invention allows for more accurate and reliable analysis. Problems, configurations, and advantages other than those described above will become clear from the description of the embodiments below.

FIG. 1 is a diagram illustrating an example of the configuration of a social participation analysis system in first and second embodiments. FIG. 11 is a diagram illustrating an example of a configuration of a user management information database used in the first and second embodiments. FIG. 2 is a diagram illustrating an example of a configuration of a social participation information database used in the first and second embodiments. FIG. 11 is a diagram illustrating an example of a configuration of an environmental information database used in the first and second embodiments. FIG. 13 is a diagram illustrating an example of a configuration of a correction amount database used in the first and second embodiments. FIG. 11 is a diagram showing an example of the configuration of a corrected social participation information database used in the first and second embodiments. FIG. 13 is a diagram showing an example of the configuration of an intervention allocation information database used in the first and second embodiments. FIG. 11 is a diagram showing an example of the configuration of a nursing care certification information database used in the first and second embodiments. FIG. 13 is a sequence diagram showing an example of social participation information analysis in the first and second embodiments. 11 is a flowchart illustrating an example of a regression model generation process in the first embodiment. 11 is a flowchart illustrating an example of a calculation process of a correction amount in the first and second embodiments. FIG. 13 is a diagram showing an example of a display screen of the analysis result in step S15 in the first and second embodiments. FIG. 13 is a diagram showing an example of a display screen of the analysis result in step S19 in the first and second embodiments. FIG. 13 is a diagram showing another example of the display screen of the analysis result in step S19 in the first and second embodiments. 13 is a flowchart illustrating an example of a regression model generation process according to the second embodiment. FIG. 11 is a functional block diagram of a social participation behavior analysis server in the first and second embodiments.

Below, examples 1 and 2, which specifically illustrate the embodiment of the present invention, are explained with reference to the drawings.

Figure 1 is a diagram showing an example of the configuration of a social participation analysis system in Examples 1 and 2. The social participation analysis system is composed of an environmental information distribution server 101, a user terminal 201, a user terminal 251, and a social participation behavior analysis server 301. The environmental information distribution server 101, the user terminal 201, the user terminal 251, and the social participation behavior analysis server 301 are connected to each other via a network 10. First, an overview of the processing of the social participation analysis system will be described. In the social participation analysis system, social participation information indicating the social participation behavior of the user is identified from sensor information acquired from other devices. Then, in the social participation analysis system, this is corrected based on environmental information at the time of the user's behavior indicated by the sensor information. For this purpose, the environmental information distribution server 101 transmits environmental information to the social participation behavior analysis server 301. In addition, the user terminal 201 transmits sensor information to the social participation behavior analysis server 301.

Then, the social participation behavior analysis server 301 identifies and corrects the social participation information. The social participation behavior analysis server 301 also analyzes the social participation information. The social participation behavior analysis server 301 also transmits the analysis results and the social participation information to other devices. As a result, the user terminal 201 and the user terminal 251 receive the analysis results on the social participation information from the social participation behavior analysis server 301 and output them. Output includes displaying on a screen. In other words, the social participation behavior analysis server 301 provides a social participation information analysis service to the user terminal 201 and the user terminal 251. The connection method of the network 10 used to realize the social participation information analysis service may be either wired or wireless, and the network 10 may be realized by the Internet, etc.

Next, the configuration of each device of the social participation analysis system will be described. First, the environmental information distribution server 101 is a server that distributes environmental information such as meteorological observation information operated by the national or local government. Meteorological observation information includes temperature, humidity, weather, etc. Note that environmental information is information that indicates the environment when social participation behavior is performed, and meteorological observation information is an example of this. For this reason, environmental information also includes location information such as location, region, and altitude, as well as indoor and outdoor conditions. This is because the same meteorological observation information affects users differently depending on the region. For example, users in cold regions are prevented from decreasing their activity to some extent even in low temperatures. In addition, the environmental information distribution server 101 can be realized by a computer having a calculation device 102, a main memory device 103, an auxiliary memory device 104, and a communication interface (communication I/F) 106.

The components of the environmental information distribution server 101 are described below. First, the arithmetic device 102 is a device that loads a program stored in the auxiliary storage device 104 into the main storage device 103 and executes it. For this reason, the arithmetic device 102 can be realized by a processor such as a CPU. In addition, the main storage device 103 temporarily stores (loads) the program executed by the arithmetic device 102 and data used when executing the program.

The auxiliary storage device 104 can be realized by a large-capacity, non-volatile storage device such as a magnetic storage device (HDD) or a semiconductor storage device (SSD) consisting of a flash memory, and stores the programs executed by the calculation device 102 and various information. Specifically, the auxiliary storage device 104 stores the programs to be loaded into the main storage device 103 and an environmental information database 105. Here, the environmental information database 105 is a database that stores environmental information such as temperature, snowfall, and precipitation for each measurement area. The communication I/F 106 is a communication interface device that controls communication with the social participation behavior analysis server 301 via a network in accordance with a specified protocol and transmits and receives data to and from the social participation behavior analysis server 301.

Next, the user terminal 201 can be realized by a terminal (such as a smartphone terminal or a wearable terminal) that measures data, and is used by the user. For this reason, it is desirable for the user terminal 201 to be a portable terminal. The measured data includes sensor information indicating the user's behavior. The user's behavior includes the movement of the user. The user terminal 201 can also be realized by a calculator (computer) having an input/output device 202, a calculation device 203, a behavior information sensor 204, a communication I/F 205, a main memory device 206, and an auxiliary memory device 208.

The input/output device 202 is formed of a display device such as a touch panel, which accepts user input as a touch panel and outputs the results of the program execution in a format that is visible to the user. For example, when the input/output device 202 accepts input from the user, it displays the results of the social participation information analysis received from the social participation behavior analysis server 301. The calculation device 203 can be realized by a processor such as a CPU, and loads a program (app) stored in the auxiliary storage device 208 into the main storage device 206 and executes it. The input/output device 202 may be configured separately into an input device such as a keyboard and an output device such as a display screen.

The behavioral information sensor 204 is composed of various behavioral information sensors that measure the state of the smartphone. Specifically, it can be composed of a location information sensor and an acceleration sensor. The communication I/F 205 is a communication interface device that controls communication with the social participation behavior analysis server 301 via a network in accordance with a specific protocol and transmits and receives data to and from the social participation behavior analysis server 301.

The main memory device 206 also temporarily stores the programs executed by the arithmetic device 203 and data used when executing the programs. A sensor information collection program 207 is loaded into the main memory device 206 as a software configuration. The sensor information collection program 207 is stored in the auxiliary memory device 208. This sensor information collection program 207 causes the arithmetic device 203 to function as a sensor information collection unit. This sensor information collection unit controls the behavior information sensor 204 and collects sensor information.

The auxiliary storage device 208 can be realized by a large-capacity non-volatile storage device such as a magnetic storage device (HDD) or a semiconductor storage device (SSD) consisting of a flash memory or the like, and stores the programs executed by the calculation device 203 and various information. Specifically, the auxiliary storage device 208 stores the sensor information collection program 207 and the sensor information database 209 to be loaded into the main storage device 206.

The sensor information database 209 is a database that stores sensor information collected according to the sensor information collection program 207. Specifically, it stores time series data of latitude and longitude collected by the position information sensor of the behavior information sensor 204, and time series data of three-dimensional acceleration collected by the acceleration sensor.

Next, the user terminal 251 can be realized by a terminal (PC or tablet terminal) that displays the above-mentioned analysis results, and is used by users such as local governments and insurance companies. Compared to the user terminal 201, the user terminal 251 has limited functions to outputting the analysis results. Furthermore, the user terminal 251 may be a portable terminal like the user terminal 201, or may be a terminal that is installed in a fixed location. The user terminal 251 is a calculator (computer) having an input/output device 252, a calculation device 253, a communication I/F 254, a main memory device 255, and an auxiliary memory device 256.

The input/output device 252 is, for example, a display device such as a touch panel, and is a device that accepts user operation input and outputs the results of program execution in a format that can be viewed by the user. For example, when the input/output device 252 of the user-owned terminal accepts input from the user, it displays the results of the social participation information analysis received from the social participation behavior analysis server 301. Note that the input/output device 252 may be configured separately into an input device such as a keyboard and an output device such as a display screen.

The calculation device 253 can be realized by a processor such as a CPU, and loads a program stored in the auxiliary storage device 256 into the main storage device 255 and executes it. The communication I/F 254 is a communication interface device that controls communication with the social participation behavior analysis server 301 via a network in accordance with a predetermined protocol, and transmits and receives data to and from the social participation behavior analysis server 301.

The main memory device 255 also temporarily stores (loads) the programs executed by the arithmetic unit 253. The auxiliary memory device 256 is configured by a large-capacity non-volatile recording device such as a magnetic storage device (HDD) or a semiconductor storage device (SSD) made of flash memory or the like, and stores the programs executed by the arithmetic unit 253.

Next, the social participation behavior analysis server 301 can be realized by a calculator having an input/output device 302, a computing device 303, a communication I/F 304, a main memory device 310, and an auxiliary memory device 350. For this reason, the social participation behavior analysis server 301 can also be expressed as a social participation behavior analysis device. The configuration of the social participation behavior analysis server 301 will be described below.
First, the input/output device 302 is a device that is composed of a display device such as a touch panel, and is a device that accepts operational input and outputs the results of program execution in a visually recognizable format. The input/output device 302 may be configured separately into an input device such as a keyboard and an output device such as a display screen. Furthermore, the social participation behavior analysis server 301 may omit the input/output device 302, or may be realized by a terminal device in a separate housing.

The calculation device 303 can be realized by a processor such as a CPU, and loads a program stored in the auxiliary storage device 350 into the main storage device 310 and executes it. The communication I/F 304 controls communication with the environmental information distribution server 101, the user terminal 201, and the user terminal 251 via the network in accordance with a specific protocol. As a result, the social participation behavior analysis server 301 can send and receive data between the environmental information distribution server 101, the user terminal 201, and the user terminal 251.

The main memory 310 also temporarily stores (loads) the programs executed by the arithmetic unit 303 and data used when executing the programs. Here, the following programs are loaded into the main memory 310 as the software configuration.
Social participation information identification program 311
Model Generation Program 312
Correction amount calculation program 313
Social participation information correction program 314
Risk Prediction Model Generation Program 315
Data Analysis Program 316
Each of these programs is stored in the auxiliary storage device 350, and when the function of each program is executed by the arithmetic device 303, it is stored in the main storage device 310. As a result, the arithmetic device 303 functions as a social participation information identification unit 33, a model generation unit 34, a correction amount calculation unit 35, a social participation information correction unit 36, a risk prediction model generation unit 37, and a data analysis unit 38. In addition to the above-mentioned programs, each of these units can also be realized by hardware. Here, the function blocks of the social participation behavior analysis server 301 will be described. FIG. 15 is a function block of the social participation behavior analysis server 301 in this embodiment and embodiment 2. In FIG. 15, the social participation behavior analysis server 301 has an input unit 31, an output unit 32, a social participation information identification unit 33, a model generation unit 34, a correction amount calculation unit 35, a social participation information correction unit 36, a risk prediction model generation unit 37, and a data analysis unit 38.

Here, the input unit 31 and the output unit 32 correspond to the communication I/F 304 and the input/output device 302 in FIG. 1. Also, the social participation information identification unit 33, the model generation unit 34, and the correction amount calculation unit 35 correspond to the social participation information identification program 311, the model generation program 312, and the correction amount calculation program 313, respectively. Furthermore, the social participation information correction unit 36, the risk prediction model generation unit 37, and the data analysis unit 38 correspond to the social participation information correction program 314, the risk prediction model generation program 315, and the data analysis program 316, respectively. Each functional block will be explained below.

The social participation information identification unit 33 identifies social participation information from the sensor information from the user terminal 201. Here, the social participation information is information indicating the user's social participation behavior, and a social participation index indicating the user's involvement with society can be used as the social participation information. In addition, the social participation information identification unit 33 may extract the social participation information from the sensor information, or may calculate or generate the social participation information based on the sensor information.

The model generation unit 34 generates a regression model with the environmental information received from the environmental information distribution server 101 as an explanatory variable and the social participation information identified by the social participation information identification unit 33 as a target variable. The model generation unit 34 generates a regression model for each type of social participation behavior. The correction amount calculation unit 35 calculates the correction amount of the social participation information based on the environmental information, the social participation information, and the regression model.

The social participation information correction unit 36 also corrects the social participation information identified by the social participation information identification unit by the calculated correction amount. In other words, the social participation information correction unit calculates the corrected social participation information by correcting the social participation information using the correction amount.

The risk prediction model generation unit 37 generates a risk prediction model for calculating the probability of being certified as requiring nursing care in the future, using the nursing care certification information stored in the nursing care certification information database 358 as the objective variable and the corrected social participation information as the explanatory variable. The data analysis unit 38 also performs analysis of the social participation information based on analysis requests from the user terminal 201 and the user terminal 251. Specifically, in response to a request from the user terminal 201, the corrected social participation information calculated by the social participation information correction unit 36 is used to evaluate social participation behavior. In response to a request from the user terminal 251, the effect of the intervention is analyzed using the corrected social participation information calculated by the social participation information correction unit 36 and the intervention allocation information stored in the intervention allocation information database.

The auxiliary storage device 350 is configured with a large capacity non-volatile storage device such as a magnetic storage device (HDD) or a semiconductor storage device (SSD) made of a flash memory or the like, and stores the programs executed by the arithmetic device 303 and various information. Specifically, the auxiliary storage device 350 includes the above-mentioned programs to be loaded into the main storage device 310. The information includes the following:
User management information database 351
Social participation information database 352
Environmental information database 353
Regression Model Database 354
Correction amount database 355
Corrected Social Participation Information Database 356
Intervention allocation information database 357
Nursing care certification information database 358
Risk Prediction Model Database 359
Here, the user management information database 351 is a database that stores user management information of the social participation system, the details of which will be described later with reference to Fig. 2. Moreover, the social participation information database 352 is a database that stores social participation information identified by the social participation information identification unit, the details of which will be described later with reference to Fig. 3. Moreover, the environmental information database 353 is a database that stores environmental information received from the environmental information distribution server 101, the details of which will be described later with reference to Fig. 4.

The regression model database 354 is a database that stores the regression model generated by the model generation unit, and stores information required to restore the regression model, such as the model type, hyperparameters, and weight parameter values. The correction amount database 355 is a database that stores the correction amount calculated by the correction amount calculation unit, the details of which will be described later using FIG. 5. The corrected social participation information database 356 is a database that stores the corrected social participation information calculated by the social participation information correction unit, the details of which will be described later using FIG. 6.

Furthermore, the intervention allocation information database 357 is a database that stores intervention allocation information for interventions such as measures for social participation behavior, the details of which will be described later using FIG. 7. Furthermore, the nursing care certification information database 358 is a database that stores nursing care certification information corresponding to users, the details of which will be described later using FIG. 8. Furthermore, the risk prediction model database 359 is a database that stores the risk prediction model generated by the risk prediction model generation unit 37, and stores information required to restore the risk prediction model, such as the type of model, hyperparameters, and weight parameter values.

Next, an example of the configuration of each database will be described. The databases shown in Figures 2 to 8 are illustrated as examples of configurations in table format, but data structures other than table format, such as lists or queues, may also be used. Also, the contents of each database are shown as examples, and some information may be deleted or new information may be added.

First, FIG. 2 is a diagram showing an example of the configuration of the user management information database 351 used in this embodiment and embodiment 2. As shown in FIG. 2, the user management information database 351 records information used for user management, including a user ID 401, a name 402, a date of birth 403, a gender 404, an address 405, and an email address 406. The user ID 401 is an identification number assigned to a user. The name 402 is the personal name of the user. The date of birth 403 is the user's date of birth. The gender 404 is the user's gender. The address 405 is information about the user's place of residence. The email address 406 is the user's email address, and is the contact information when the social participation behavior analysis server 301 contacts the user by email.

Next, FIG. 3 is a diagram showing an example of the configuration of the social participation information database 352 used in this embodiment and embodiment 2. The social participation information database 352 is a database that records the social participation information of the user identified by the social participation information identification unit 33, and includes basic information 501 and measurement data 502. The basic information 501 includes a user ID 503, a measurement date 504, and a region 505. The measurement data 502 includes the number of steps 506, a movement distance 507, and a movement time 508. The measurement data 502 may include social participation behavior calculated from sensor information measured by the user terminal 251. Here, the social participation behavior includes at least one of the following. That is, the social participation information identification unit 33 identifies the social participation information exemplified below.
Number of times going out Number of places stayed Status of stay at specific spots (number of times, duration, etc.)
Car useMotorcycle useRail useBus useNumber of tripsWalking timeContinuous walking timeHigh intensity activity timeParticipation in group activities (number of times, duration, etc.)
Working status (number of times, hours, etc.)
Of these, it is preferable to use at least one of the following: the number of places the user stayed, the status of staying at a specific spot, the status of participation in group activities, and the status of employment, because these pieces of information are characteristic information about the user's behavior.

User ID 503 is an identification number assigned to the user. Measurement date 504 is the date when the social participation information was measured. Area 505 is the measurement area of the social participation information. The area division recorded in area 505 may be any area division, such as prefecture, city, ward, town, or village. Number of steps 506 is the total number of steps taken by the user in one day, extracted by the social participation information identification unit 33 from the sensor information measured by the user terminal 201. Travel distance 507 is the total travel distance of the user in one day, identified by the social participation information identification unit 33 from the sensor information measured by the user terminal 201. Travel time 508 is the total travel time of the user in one day, identified by the social participation information identification unit 33 from the sensor information measured by the user terminal 201.

Next, FIG. 4 is a diagram showing an example of the configuration of the environmental information database 353 used in this embodiment and embodiment 2. The environmental information database 353 is a database that records environmental information received from the environmental information distribution server 101, and includes basic information 601 and measurement data 602. The basic information 601 includes an area 603 and a measurement date 604, and the measurement data 602 includes an average temperature 605, an amount of snowfall 606, and an amount of precipitation 607. The measurement data 602 may also include information such as hours of sunshine, wind speed, humidity, air pressure, and amount of ultraviolet rays. The amount of snowfall may also be used instead of or in addition to the amount of snowfall 606.

Here, region 603 is the measurement region of the environmental information. The region division recorded in region 603 may be any region division, such as prefecture, city, ward, town, or village. Measurement date 604 is the date on which the environmental information was measured. Average temperature 605 is the average value of the observed temperature for the day. Snowfall amount 606 is the amount of snowfall over 24 hours. Precipitation amount 607 is the amount of precipitation over 24 hours.

Next, FIG. 5 is a diagram showing an example of the configuration of the correction amount database 355 used in this embodiment and embodiment 2. The correction amount database 355 is a database that records the correction amount of the social participation information calculated by the correction amount calculation unit 35, and includes basic information 701 and measurement data 702. The basic information 701 includes a user ID 703, a measurement date 704, and an area 705, and the measurement data 702 includes a step count correction amount 706, a movement distance correction amount 707, and a movement time correction amount 708.

The user ID 703 is an identification number assigned to the user. The measurement date 704 is the date on which the social participation information was measured. The area 705 is the area on which the social participation information was measured. The step count correction amount 706 is the correction amount for the step count 506. The movement distance correction amount 707 is the correction amount for the movement distance 507. The movement time correction amount 708 is the correction amount for the movement time 508. A specific method for calculating the correction amount will be described later with reference to FIG. 11.

Next, FIG. 6 is a diagram showing an example of the configuration of the corrected social participation information database 356 used in this embodiment and embodiment 2. The corrected social participation information database 356 is a database that records the corrected social participation information calculated by the social participation information correction unit 36, and includes basic information 801 and measurement data 802. The basic information 801 includes a user ID 803, a measurement date 804, and an area 805, and the measurement data 802 includes a corrected number of steps 806, a corrected moving distance 807, and a corrected moving time 808. Here, the user ID 803 is an identification number assigned to the user. The measurement date 804 is the date on which the social participation information was measured. The area 805 is the area where the social participation information was measured. The corrected number of steps 806 is the corrected number of steps 506. The corrected moving distance 807 is the corrected moving distance 507. The corrected moving time 808 is the corrected moving time 508. A specific method for calculating the corrected social participation information will be described later in FIG. 9.

Next, FIG. 7 is a diagram showing an example of the configuration of the intervention allocation information database 357 used in this embodiment and embodiment 2. The intervention allocation information database 357 includes a policy ID 901, a policy name 902, an area 903, a policy start date and time 904, and a user ID 905, and records intervention allocation information used for analysis in the data analysis unit 38. Here, the policy ID 901 is an identification number assigned to the intervention policy. The policy name 902 is the name of the intervention policy. The area 903 is the area where the intervention policy is implemented. The policy start date and time 904 is the start date and time of the policy, and can also be expressed as the intervention start date and time. The user ID 905 is an identification number assigned to a user, and registers the user ID of a user who participated in the policy.

Next, FIG. 8 is a diagram showing an example of the configuration of the nursing care certification information database 358 used in this embodiment and embodiment 2. The nursing care certification information database 358 records nursing care certification information including a user ID 1001, a name 1002, a date of birth 1003, a gender 1004, an insured person number 1005, a certification date 1006, and a nursing care certification status category 1007. The nursing care certification information is used for analysis in the risk prediction model generation unit 37. The nursing care status category 1007 can also be expressed as a nursing care certification category.

Here, user ID 1001 is an identification number assigned to a user, and the user ID of a user who has received nursing care certification is registered. Name 1002 is the user's personal name. Date of birth 1003 is the user's date of birth. Gender 1004 is the user's gender. Insured person number 1005 is the user's insured person number. Certification date 1006 is the date on which the user was certified in nursing care status category 1007. Nursing care status category 1007 is the user's nursing care status category, and its elements are nursing care 1, nursing care 2, nursing care 3, nursing care 4, and nursing care 5. Support required 1 and support required 2 may also be added as elements.

Furthermore, the same user is assigned the same user ID 401, user ID 503, user ID 703, user ID 803, user ID 905, and user ID 1001. Therefore, it is possible to perform the following search using user ID 401, user ID 503, user ID 703, user ID 803, user ID 905, and user ID 1001. That is, the following information is searched from the user management information database 351, social participation information database 352, correction amount database 355, corrected social participation information database 356, intervention allocation information database 357, and nursing care certification information database 358.
User management information of the same user Social participation information Correction amount Corrected social participation information Intervention allocation information Nursing care certification information This concludes the explanation of the database, and next, the processing in this embodiment will be explained. Fig. 9 is a sequence diagram showing an example of social participation information analysis in this embodiment. The processing of each step in Fig. 9 will be explained below. In the following explanation, the parts shown in Fig. 15 are used as the processing subjects in the processing in the social participation behavior analysis server 301.

(Figure 9: Step S1)
First, the user terminal 201 accepts input of a user registration application from the user via the input/output device 202. Then, the user terminal 201 transmits user application information corresponding to the accepted user registration application to the social participation behavior analysis server 301 via the communication I/F 205.

(FIG. 9: Step S2)
Next, the input unit 31 of the social participation behavior analysis server 301 receives the user registration application information from the user terminal 201. Furthermore, the social participation information identification unit 33 and the like update the user management information stored in the user management information database 351 in accordance with the user registration application information. Then, the output unit 32 notifies the user terminal 201 that registration is complete. Specifically, the user management information is updated by assigning a user ID 401 to the user registration application information, adding an entry to the user management information database 351, and registering the name 402, date of birth 403, gender 404, address 405, and email address 406.

(Figure 9: Step S3)
Furthermore, the user terminal 201 receives a registration completion notification via the communication I/F 205. Furthermore, on condition that the registration completion notification is received or the sensor information collection program 207 is started, the calculation device 203 starts measuring sensor information in accordance with the sensor information collection program 207. Then, the calculation device 203 stores the measured sensor information in the sensor information database 209 in accordance with the sensor information collection program 207. Specifically, the calculation device 203 controls the behavior information sensor 204 in accordance with the sensor information collection program 207 to measure sensor information and registers the measured sensor information in the sensor information database 209.

(Figure 9: Step S4)
The user terminal 201 transmits the sensor information stored in the sensor information database 209 to the social participation behavior analysis server 301 at any timing. Specifically, the computing device 203 periodically extracts the sensor information in accordance with the sensor information collection program 207 and transmits it via the communication I/F 205. Periodic transmission includes, for example, transmitting the sensor information at the same time every day.

(FIG. 9: Step S5)
The input unit 31 of the social participation behavior analysis server 301 receives sensor information from the user terminal 201. The social participation information identification unit 33 identifies the social participation information from the received sensor information. The social participation information identification unit 33 updates the social participation information database 352 with the identified social participation information. Specifically, the social participation information identification unit 33 extracts the number of steps, the movement distance, and the movement time from the acceleration information and the position information included in the sensor information. Next, the social participation information identification unit 33 adds an entry to the social participation information database 352, and registers the extracted items, the user ID 503, the measurement date 504, the area 505, the number of steps 506, the movement distance 507, and the movement time 508. Note that any method can be used as a method for determining the area 505 to be registered in the social participation information database 352. For example, a method of using an address calculated from the center of gravity of the position information measured by the user terminal 201, or a method of using the address 405 by referring to the user management information database 351 can be considered.

The identification of the social participation information in step S5 may be performed as follows. First, the user terminal 201 may identify the social participation information, and then transmit the social participation information from the user terminal 201 to the social participation behavior analysis server 301.

(FIG. 9: Step S6)
Furthermore, the social participation behavior analysis server 301 requests environmental information from the environmental information distribution server 101. For this purpose, the output unit 32 periodically transmits an environmental information request to the environmental information distribution server 101. Periodic transmission includes, for example, transmission at the same time every day. In this way, it is desirable to match the period of step S4 with that of this step.

(Figure 9: Step S7)
Furthermore, the communication I/F 106 of the environmental information distribution server 101 receives an environmental information request from the social participation behavior analysis server 301. Then, the computing device 102 reads environmental information corresponding to the environmental information request from the environmental information database 105. As a result, the communication I/F 106 transmits the read environmental information to the social participation behavior analysis server 301.

As a result, the input unit 31 of the social participation behavior analysis server 301 receives the environmental information. The model generation unit 34 then updates the environmental information database 353 in accordance with the received environmental information. Specifically, the model generation unit 34 adds an entry to the environmental information database 353, and registers the region 603, measurement date 604, average temperature 605, snowfall amount 606, and precipitation amount 607 based on the received environmental information.

(Figure 9: Step S8)
Next, the model generation unit 34 of the social participation behavior analysis server 301 generates a regression model at an arbitrary timing. This regression model is created for each type of corresponding social participation information. Then, the model generation unit 34 updates the regression model database 354 according to the created regression model. The arbitrary timing includes, for example, a certain period of time after the previous execution of model generation. Hereinafter, the details of the generation of this regression model will be described with reference to FIG. 10. FIG. 10 is a flowchart for explaining an example of the generation process of the regression model in this embodiment.

(FIG. 10: Step S20)
The model generation unit 34 reads out the target social participation information from the social participation information database 352. For example, the social participation information stored in the social participation information database 352 after the previous execution of model generation is read out.
(FIG. 10: Step S21)
The model generation unit 34 reads out environmental information corresponding to the social participation information read out in step S21 from the environmental information database 353. Here, the environmental information corresponding to the social participation information in this embodiment refers to environmental information in which at least one of the respective regions (505 or 603) and the measurement date (504 or 604) matches. The order of processing steps S20 and S21 does not matter. If step S21 is performed first, in step S20, social participation information corresponding to the environmental information read out according to a predetermined condition is read out. Here, the predetermined condition includes having been registered since the previous execution of model generation. Furthermore, the model generation unit 34 may specify a predetermined condition and execute steps S20 and S21 accordingly.

(FIG. 10: Step S22)
Next, the model generation unit 34 combines the read social participation information with the environmental information. Specifically, the model generation unit 34 identifies the environmental information using the measurement date 504 (corresponding to the measurement date 604) and the region 505 (corresponding to the region 603) of the social participation information as keys, and performs an internal combination of the social participation information and the identified environmental information to generate combined data. Note that in step S20 or step S21, if there is only one type of measurement date or region, it is not necessary to identify the environmental information using this as a key. Also, the model generation unit 34 may identify the social participation information using the measurement date or region of the environmental information as a key. Furthermore, the identification of the corresponding social participation information or environmental information in step S20 or step S21 may be omitted.

In this step, an internal join is performed with the source being environmental information and the destination being social participation information, so the number of entries in the joined data will be the same as the number of entries in the social participation information read out. The columns of the joined data will be user ID 503, measurement date 504, area 505, number of steps 506, distance traveled 507, travel time 508, average temperature 605, amount of snowfall 606, and amount of precipitation 607.

(FIG. 10: step S23)
Furthermore, the model generation unit 34 uses the combined data to generate a regression model of the social participation information against the environmental information for each type of social participation behavior. Specifically, the model generation unit 34 uses the combined data generated in step S22 to create a regression model for each type of social participation behavior shown in the following (1) to (3).
(1) A regression model in which the objective variable is the number of steps 506 and the explanatory variables are the average temperature 605, the amount of snowfall 606, and the amount of precipitation 607;
(2) A regression model in which the objective variable is the travel distance 507 and the explanatory variables are the average temperature 605, the amount of snowfall 606, and the amount of precipitation 607;
(3) A regression model in which the objective variable is travel time 508 and the explanatory variables are average temperature 605, snowfall amount 606, and precipitation amount 607.

Any type of regression model can be used. For example, a linear regression model, a neural network, a random forest, a support vector machine, or a gradient boosting tree can be used. In this way, by generating a regression model for each type of social participation information based on measurement data, it becomes possible to quantitatively and objectively grasp the unknown relationship between environmental information and each type of social participation information based on the data. As a result, it is expected that the accuracy and reliability of the analysis will improve.

In addition, information that influences people's social participation behavior may be added as an explanatory variable of the regression model. Such information includes basic information about the user, such as age, gender, and biometric information, calendar information such as whether the measurement date is a weekday, weekend, or holiday, and the number of new infections and serious patients in the event of an infectious disease pandemic. Also, taking into account that social participation behavior differs by region, a regression model may be generated for each region based on region 505 and used in subsequent analyses.

(FIG. 10: Step S24)
Furthermore, the model generating unit 34 updates the regression model database 354 based on the generated regression model. This completes the explanation of Fig. 10, and we will return to the explanation of Fig. 9.

(FIG. 9: Step S9)
Furthermore, the correction amount calculation unit 35 can execute the correction amount calculation of the social participation information at any timing, regardless of the execution timing of the generation of the regression model (step S8), and update the correction amount database 355. The correction amount calculation unit 35 executes the correction amount calculation periodically, for example, at the same time every day, and updates the correction amount database 355. Here, the details of the correction amount calculation in this step will be described with reference to FIG. 11. FIG. 11 is a flowchart for explaining an example of the calculation process of the correction amount in this embodiment and embodiment 2.

(FIG. 11: Step S30)
The correction amount calculation unit 35 reads out a regression model from the regression model database 354 for each created social participation behavior.

(FIG. 11: Step S31)
The correction amount calculation unit 35 substitutes the reference environmental information (hereinafter, reference environmental information) into the read regression model to calculate the output. Specifically, the correction amount calculation unit 35 calculates the outputs S0, D0, and T0 according to (Equation 1), (Equation 2), and (Equation 3), respectively. In (Equation 1), (Equation 2), and (Equation 3), f is a function of a regression model with the number of steps 506 as the objective variable, g is a function of a regression model with the travel distance 507 as the objective variable, and h is a function of a regression model with the travel time 508 as the objective variable. X0, Y0, and Z0 are reference environmental information, and arbitrary values of the average temperature 605, snowfall amount 606, and precipitation amount 607 are set, respectively. For example, X0=20° C., Y0=0 mm, and Z0=0 mm are set.

S0 = f(X0, Y0, Z0) ... (Equation 1)
D0 = g(X0, Y0, Z0) ... (Equation 2)
T0 = h(X0,Y0,Z0)...(Math. 3)
(FIG. 11: Step S32)
The correction amount calculation unit 35 reads out from the social participation information database 352 the social participation information that has been newly added since the previous correction amount calculation was performed.

(FIG. 11: step S33)
The correction amount calculation unit 35 reads out environmental information corresponding to the social participation information read out in step S32 from the environmental information database 353. The order of processing steps S32 and S33 does not matter. If step S33 is performed first, social participation information corresponding to the environmental information read out according to a predetermined condition is read out in step S32.

(FIG. 11: Step S34)
Since this is the same process as step S22, the explanation will be omitted.

(FIG. 11: Step S35)
The correction amount calculation unit 35 substitutes the combined data obtained in step S34 into the read regression model and calculates the output. Specifically, the correction amount calculation unit 35 calculates the outputs S, D, and T according to (Equation 4), (Equation 5), and (Equation 6). In (Equation 4), (Equation 5), and (Equation 6), X, Y, and Z are observed environmental information, X is the average temperature 605, Y is the amount of snowfall 606, and Z is the amount of precipitation 607.

S = f(X,Y,Z)...(Math. 4)
D = g(X,Y,Z)...(Math. 5)
T = h(X,Y,Z)...(Math. 6)
(FIG. 11: Step S36)
The correction amount calculation unit 35 calculates the correction amount based on the processing results of steps S31 and S35. Specifically, the correction amount calculation unit 35 calculates the correction amount based on the processing results of steps S31 and S35 using Equation 7, Equation 8, and Equation 9. According to (Equation 9), a step number correction amount 706 (ΔS), a movement distance correction amount 707 (ΔD), and a movement time correction amount 708 (ΔT) are calculated.

ΔS = S0 - S ... (Math. 7)
ΔD = D0 - D...(Math. 8)
ΔT = T0 - T...(Equation 9)
(FIG. 11: Step S37)
The correction amount calculation unit 35 updates the correction amount database 355 based on the correction amount calculated in step S36. Specifically, the correction amount calculation unit 35 adds an entry to the correction amount database 355 and updates the user ID 703. The measurement date 704, the area 705, the step count correction amount 706, the movement distance correction amount 707, and the movement time correction amount 708 are registered. In addition, the model generation (step S8) is executed, and the initial correction amount calculation (step S9) is executed. In this case, the correction amount calculation unit 35 may execute a correction amount calculation for each piece of social participation information registered in the social participation information database 352. Each registered entry will be updated. This update may be applied to all social participation information (entries).

The correction amount is calculated as the difference between the predicted value (S0, D0, T0) of the social participation information measured under the reference environmental information and the predicted value (S, D, T) of the social participation information measured under the actually observed environmental information, as shown in (Equation 7), (Equation 8), and (Equation 9). In other words, the correction amount can also be expressed as a quantity that represents the impact that the difference between the reference environmental information and the actual environmental information has on the social participation information.

The calculation of the correction amount above involves setting a predetermined standard (absolute standard) and correcting the social participation information for at least one period. Furthermore, rather than using an absolute standard, a correction may be made such that the social participation information for a first period out of a number of predetermined periods (for example, cycles) corresponds to the social participation information for a second period. In other words, a relative correction may be made using the social participation information for the second period as the standard. Furthermore, the first period and the second period each indicate a period (cycle) either before or after the intervention, but either may be the previous period or the subsequent period. This concludes the explanation of FIG. 11, and we return to the explanation of FIG. 9.

(FIG. 9: Step S10)
The social participation information correction unit 36 corrects the social participation information using the correction amount. Then, the social participation information correction unit 36 updates the corrected social participation information database 356. Specifically, the social participation information correction unit 36 reads out the social participation information from the social participation information database 352, and reads out the correction amount from the correction amount database 355. Next, the social participation information correction unit 36 associates the read out social participation information and the correction amount with the user ID 503 (corresponding to the user ID 703) and the measurement date 504 (corresponding to the measurement date 704) as keys. Then, the social participation information correction unit 36 calculates the sum of the number of steps 506 and the step number correction amount 706 for each associated user ID 503 and measurement date 504, and calculates the corrected number of steps 806.

Similarly, the social participation information correction unit 36 sums the movement distance 507 and the movement distance correction amount 707 to calculate the corrected movement distance 807. Similarly, the social participation information correction unit 36 sums the movement time 508 and the movement time correction amount 708 to calculate the corrected movement time 808. As described above, the social participation information correction unit 36 calculates the corrected social participation information for each type of social participation behavior.

Then, the social participation information correction unit 36 adds an entry to the corrected social participation information database 356 and registers the user ID 803, measurement date 803, area 805, corrected number of steps 806, corrected movement distance 807, and corrected movement time 808. For example, in the case of an entry in which the user ID 503 is u001 and the measurement date 504 is 2021/11/1, the number of steps 506 is 8000 steps and the step number correction amount 706 is +200 steps, so the social participation information correction unit 36 registers 8200 steps in the corrected number of steps 806. Similarly, the movement distance 507 is 15 km and the movement distance correction amount 707 is +1.2 km, so the social participation information correction unit 36 registers 16.2 km in the corrected movement distance 807. Additionally, since the travel time 508 is 120 minutes and the travel time correction amount 708 is +12 minutes, the social participation information correction unit 36 registers 132 minutes as the corrected travel time 808.

In this way, the social participation information correction unit 36 corrects the social participation information based on the environmental information. This makes it possible to perform analysis using data on social participation information with different measurement environments. For example, data measured at different times or in different locations will have significantly different environments at the time of measurement depending on the data, but by applying correction, it becomes possible to compare data measured in different environments, leading to improved analytical accuracy and reliability.

(FIG. 9: Step S11)
The above processing up to step S10 concludes the identification and correction of the social participation information. Below, an analysis is performed using at least the corrected social participation information, but it is assumed that this is performed at a different timing from the identification and correction of the social participation information. However, the timing is not limited. Also, the following processing may be omitted.

The risk prediction model generation unit 37 generates a risk prediction model at any timing. Then, the risk prediction model generation unit 37 updates the risk prediction model database 359 with the generated risk prediction model. For example, the risk prediction model generation unit 37 acquires new nursing care certification information and executes risk prediction model generation when the nursing care certification information database 358 is updated. Details of risk prediction model generation are described below.

First, the risk prediction model generation unit 37 reads out corrected social participation information from the corrected social participation information database 356. The risk prediction model generation unit 37 also reads out nursing care certification information from the nursing care certification information database 358. The risk prediction model generation unit 37 then constructs a nursing care risk prediction model using the corrected social participation information and the nursing care certification information. Specifically, the risk prediction model generation unit 37 generates a risk prediction model using the nursing care state classification 1007 as the objective variable and the corrected number of steps 806, the corrected movement distance 807, and the corrected movement time 808 as explanatory variables.

The risk prediction model generation unit 37 also updates the risk prediction model database 359 based on the generated risk prediction model. The risk prediction model generated here returns the probability of being certified as requiring nursing care within three years when the corrected number of steps 806, corrected movement distance 807, and corrected movement time 808 are substituted. The prediction period can be set arbitrarily. Furthermore, any classification model can be used as the risk prediction model, such as logistic regression, neural network, or random forest.

In addition, information such as the user's age, sex, biological information, medication information, and medical history may be added as explanatory variables for these classification models. Furthermore, other risk prediction models, such as a mortality risk prediction model and a disease incidence risk prediction model, may be constructed by using mortality information, disease incidence information, and the like as objective variables for the classification models.

(FIG. 9: Step S12)
When the user terminal 251 receives an input of an analysis request from a user via the input/output device 252, the user terminal 251 transmits analysis request information to the social participation behavior analysis server 301 via the communication I/F 254. The analysis request information includes a policy ID 901 corresponding to a policy for which the user wants to examine the effect of intervention.

(FIG. 9: Step S13)
When the data analysis unit 38 receives the data in step S12, it executes data analysis. The data analysis will be described in detail below. First, the data analysis unit 38 reads out corrected social participation information corresponding to the analysis request information from the corrected social participation information database 356 and intervention allocation information from the intervention allocation information database 357. The corrected social participation information and intervention allocation information corresponding to the analysis request information refer to, for example, information corresponding to the policy ID 901 of the analysis request information.

The data analysis unit 38 also extracts, from the read corrected social participation information, the corrected social participation information measured before the intervention and the corrected social participation information measured after the intervention. Specifically, the data analysis unit 38 associates the corrected social participation information with the intervention allocation information using the user ID 803 (corresponding to the user ID 905) as a key. Then, the data analysis unit 38 extracts, from the entries having the policy ID 901 to be analyzed in a field, entries whose measurement date 804 is before the policy start date/time 904 and entries whose measurement date 804 is after the policy start date/time 904.

The pre-intervention period and post-intervention period to be analyzed may be any period. For example, the pre-intervention period is set to 28 days before the start of the intervention, and the post-intervention period is set to 28 days after the start of the intervention.

Next, the data analysis unit 38 reads out the risk prediction model from the risk prediction model database 359. The data analysis unit 38 also substitutes the corrected social participation information measured before the start of the intervention and the corrected social participation information measured after the start of the intervention into the risk prediction model. The data analysis unit 38 then calculates the risk score before the start of the intervention and the risk score after the start of the intervention, respectively. Specifically, the data analysis unit 38 calculates the risk score before the start of the intervention Rpre and the risk score after the start of the intervention Rpost according to the following (Equation 7) and (Equation 8). In (Equation 7) and (Equation 8), r indicates a function of the risk prediction model. Also, Spre indicates the corrected number of steps 806 in the period before the start of the intervention. Also, Dpre indicates the corrected moving distance 807 in the period before the start of the intervention. Also, Tpre indicates the corrected moving time 808 in the period before the start of the intervention. Also, Spost indicates the corrected number of steps 806 in the period after the start of the intervention. Also, Dpost indicates the corrected moving distance 807 in the period after the start of the intervention. Furthermore, Tpost indicates the corrected movement time 808 after the start of the intervention.

Rpre = r(Spre,Dpre,Tpre)...(Math. 7)
Rpost = r(Spost, Dpost, Tpost) ... (Equation 8)
In addition, the data analysis unit 38 uses the calculated risk scores to calculate, for each user, an average value of the risk scores for the period before the start of the intervention and an average value of the risk scores for the period after the start of the intervention. The data analysis unit 38 may calculate the risk score based on the social participation information before correction, instead of the corrected social participation information. The risk score is calculated by substituting the above. First, the number of steps 506 in the period before the intervention is started is substituted into Spre. The distance traveled 507 in the period before the intervention is started is substituted into Dpre. The movement time 508 during the post-intervention period is substituted into Spost. The number of steps 506 during the post-intervention period is substituted into Dpost. The movement distance 507 during the post-intervention period is substituted into Tpost. Substitute time 508.

(FIG. 9: Step S14)
When step S13 is completed, the output unit 32 transmits the analysis results to the user terminal 251. Specifically, the output unit 32 transmits the risk score calculated in step S13 and other information to be displayed on the screen of the user terminal 251 to the user terminal 251. The other information will be described with reference to FIG. 12 in the explanation of step S15.

(FIG. 9: Step S15)
When the user terminal 251 receives the data analysis result, it displays the analysis result on the input/output device 252. FIG. 12 is a diagram showing an example of a display screen of the analysis result in this step, and shows the analysis result of the intervention effect. FIG. 12 will be described below. This display screen of the analysis result includes an environmental information display area 1201 and an analysis result display area 1204. First, the environmental information display area 1201 includes an environmental information table display area 1202 and a correction amount display area 1203. Furthermore, the environmental information table display area 1202 displays the average values of the temperature, snowfall, and precipitation in the pre-intervention period, and the average values of the temperature, snowfall, and precipitation in the post-intervention period. Specifically, in step S14, the following various information received from the social participation behavior analysis server 301 is displayed.
The average values of the average temperature 605, snowfall 606, and precipitation 607 in the pre-intervention period, and the average values of the average temperature 605, snowfall 606, and precipitation 607 in the post-intervention period.

Furthermore, the influence of the difference in environmental information between the pre-intervention period and the post-intervention period on the social participation information is displayed in the correction amount display area 1203. Specifically, in step S14, the following various information received from the social participation behavior analysis server 301 is displayed.
Difference between the average value of the step count correction amount 706 in the pre-intervention period and the average value of the step count correction amount 706 in the post-intervention period Difference between the average value of the movement distance correction amount 707 in the pre-intervention period and the average value of the movement distance correction amount 707 in the post-intervention period Difference between the average value of the movement time correction amount 708 in the pre-intervention period and the average value of the movement time correction amount 708 in the post-intervention period In addition, the analysis result display area 1204 includes a correction yes/no selection section 1205 and an intervention effect display area 1206. First, the correction yes/no selection section 1205 is used to select whether the analysis result to be displayed in the intervention effect display area 1206 is calculated based on the social participation information after correction or based on the social participation information before correction. Specifically, when the user selects "with correction" in the correction yes/no selection section 1205, in step S13, the risk score calculated by substituting the corrected step count 806, the corrected movement distance 807, and the corrected movement time 808 is displayed. Moreover, when "no correction" is selected in the correction/non-correction selection unit 1205, in step S13, the number of steps 506, the movement distance 507, and the movement time 508 are substituted and the calculated risk score is displayed.

The intervention effect display area 1206 also displays the risk score for the pre-intervention period and the risk score for the post-intervention period calculated in step S13. Specifically, the intervention effect display area 1206 displays a box plot and numerical values comparing the risk score for the pre-intervention period and the risk score for the post-intervention period. In addition, the intervention effect display area 1206 displays a P-value calculated by comparing the risk score for the pre-intervention period and the risk score for the post-intervention period using a statistical test.

In the intervention effect display area 1206, indices other than the risk score may be displayed as indices used for comparison before and after the intervention. For example, the number of steps, distance traveled, and travel time may be used as comparison indices. The above data analysis was performed in response to a request from the user terminal 251, but it is also possible to perform it in response to a request from the user terminal 201. The details are explained below.

(FIG. 9: Step S16)
When the user terminal 201 receives an analysis request input from the user via the input/output device 252, it transmits analysis request information to the social participation behavior analysis server 301 via the communication I/F 205. In other words, it executes the same process as in step S12.

(FIG. 9: Step S17)
When the data analysis unit 38 receives the analysis request information from the user terminal 201 in step S16, the data analysis unit 38 executes the data analysis. Details of the data analysis in this step are explained below. Note that the analysis in this step is a process equivalent to step S13 when the analysis request information is received from the user terminal 251.

First, the data analysis unit 38 reads out corrected social participation information corresponding to the analysis request information from the corrected social participation information database 356. Here, the analysis request information includes a user ID that identifies the user of the user terminal 201. The user ID may be received as information separate from the analysis request information. The data analysis unit 38 evaluates the social participation behavior of the user based on the read corrected social participation information, that is, the corrected social participation information corresponding to the user ID of the analysis request information. Specifically, the data analysis unit 38 evaluates the corrected number of steps 806, corrected movement distance 807, and corrected movement time 808 of the user based on arbitrarily set thresholds. For example, the rank is determined according to the corrected number of steps 806. Assume that the thresholds used to evaluate the corrected number of steps 806 are set to 3000 steps and 7000 steps. In this case, if the user's corrected number of steps 806 is less than 3000 steps, the user is evaluated as rank C; if the user's corrected number of steps 806 is 3000 steps or more but less than 7000 steps, the user is evaluated as rank B; and if the user's corrected number of steps 806 is 7000 steps or more, the user is evaluated as rank A. In addition, the corrected movement distance 807 and the corrected movement time 808 can also be evaluated in a similar manner. The number and values of the thresholds used for the evaluation may be set arbitrarily. Furthermore, the method of expressing the evaluation may be replaced with arbitrary words.

(FIG. 9: Step S18)
When step S17 is completed, the output unit 32 transmits the analysis results to the user terminal 201. Specifically, the output unit 32 transmits the evaluation of the user's social participation information calculated in step S17 and other information to be displayed on the screen of the user terminal 201 to the user terminal 201. The other information will be described with reference to Figures 13A and 13B in the explanation of step S19.

(FIG. 9: Step S19)
When the user terminal 201 receives the data analysis result, it displays the analysis result on the input/output device 202. Figures 13A and 13B are diagrams showing the display screen of the analysis result in this step. First, Figure 13A will be described.

The display screen shown in FIG. 13A includes a correction selection section 1301, a social participation evaluation display area 1302, and a correction explanation display area 1303. First, the correction selection section 1301 is used to select whether the analysis results to be displayed in the social participation evaluation display area 1302 are calculated based on the social participation information after correction or based on the social participation information before correction. Specifically, if the user selects "with correction" in the correction selection section 1301, the evaluation result based on the corrected social participation information read from the corrected social participation information database 356 is displayed in step S17. Also, if "no correction" is selected in the correction selection section 1301, the evaluation result based on the social participation information read from the social participation information database 352 is displayed in step S17. The following description will explain the display screen when "with correction" is selected in the correction selection section 1301.

The social participation evaluation display area 1302 displays the analysis results received from the social participation behavior analysis server 301 in step S18. Specifically, the evaluation calculated in step S17 and the corrected social participation information stored in the corrected social participation information database 356 are displayed on the screen.

The correction explanation display area 1303 is used to display an explanation regarding the correction, and when the user selects it, the display screen switches to Figure 13B.

The following describes FIG. 13B, which is a display screen showing the analysis results in this step. The display screen shown in FIG. 13B includes an environmental information display area 1304 and a corrected social participation information display area 1305. First, the environmental information display area 1304 displays the environmental information for the measurement period received in step S18. Specifically, the average values for the average temperature 605, snowfall 606, and precipitation 607 for the measurement period are displayed.

The corrected social participation information display area 1305 displays the corrected social participation information and the correction amount received in step S18. Specifically, the corrected number of steps 806, corrected distance traveled 807, corrected travel time 808, the average values during the measurement period of the step number correction amount 706, travel distance correction amount 707, and travel time correction amount 708 are displayed. This concludes the explanation of Example 1.

Next, Example 2 will be described. In this example, model generation is performed based on the degree of fit of the model to the data in the model generation (step S8) of the social participation behavior analysis server 301 in Example 1. The difference from Example 1 is the model generation (step S8) in the model generation unit 34 of the social participation behavior analysis server 301. Therefore, step S8 will be described below with reference to FIG. 14. FIG. 14 is a flowchart explaining an example of the regression model generation process in Example 2.

(FIG. 14: Steps S40 to S42)
Since this is the same process as steps S20 to S22 in FIG. 10, the description will be omitted.

(FIG. 14: step S43)
The model generating unit 34 reads out the regression model from the regression model database 354. At this time, if the regression model does not exist in the regression model database 354, step S46 is executed.

(FIG. 14: Step S44)
The model generation unit 34 also calculates the degree of fit of the model to the data using the combined data generated in step S42 and the regression model read out in step S43. Specifically, the model generation unit 34 substitutes the average temperature 605, snowfall amount 606, and precipitation amount 607 of the combined data into a regression model with the number of steps 506 as the objective variable to calculate the output. Next, the model generation unit 34 calculates the degree of fit of the model to the data using the output of the model and the number of steps 506. For example, the coefficient of determination or Root Mean Squared Error (RMSE) is used as an index of the degree of fit of the model to the data. The model generation unit 34 also performs the same process on the regression model with the travel distance 507 and travel time 508 as the objective variables, and calculates the degree of fit of the model to the data for each model.

(FIG. 13: Step S45)
The model generation unit 34 also determines whether to update the model based on the degree of fit calculated in step S44. Specifically, if the degree of fit calculated in step S44 is higher than a preset criterion (YES), the model generation unit 34 ends the process. If the degree of fit is lower than a preset criterion (NO), the model generation unit 34 transitions to step S46. That is, model generation is executed.

(FIG. 13: Steps S46 and S47)
Since these are the same processes as steps S23 and S24 in FIG. 10, respectively, the explanation will be omitted.

In this way, by calculating the degree of fit of the model to the data, it is possible to update the model when the relationship between the environmental information and each piece of social participation information changes. This makes it possible to always use a regression model that accurately models the relationship between the environmental information and each piece of social participation information, leading to improved analytical precision and reliability. Furthermore, the model update process is only performed when the relationship between the environmental information and each piece of social participation information changes. Therefore, the amount of calculation required is smaller than when the model generation process is performed periodically, leading to a reduced load on the social participation behavior analysis server 301.

The present invention is not limited to the above-described embodiments, but includes various modified examples. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those having all of the configurations described. In addition, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment.

It is also possible to add, delete, or replace part of the configuration of each embodiment with other configurations. The above configurations, functions, processing units, processing means, etc. may be realized in hardware, in part or in whole, for example by designing them as integrated circuits. The above configurations, functions, etc. may also be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the programs, tables, and files that realize each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

101...Environmental information distribution server, 201...User terminal, 251...User terminal, 301...Social participation behavior analysis server, 31...Input unit, 32...Output unit, 33...Social participation information identification unit, 34...Model generation unit, 35...Correction amount calculation unit, 36...Social participation information correction unit, 37...Risk prediction model generation unit, 38...Data analysis unit, 352...Social participation information database, 353...Environmental information database, 354...Regression model database, 355...Correction amount database, 356...Correction social participation information database

Claims (14)

A social participation behavior analysis device that analyzes a user's social participation behavior,
an input unit that receives sensor information indicating actions of a plurality of users and environmental information indicating an environment at the time of the actions;
a social participation information identification unit that identifies social participation information indicating a social participation behavior of the user from the sensor information;
a model generation unit that generates a regression model of the social participation information with respect to the environmental information for each type of social participation behavior;
a correction amount calculation unit that calculates a correction amount of the social participation information corresponding to the environmental information based on the regression model;
A social participation behavior analysis device comprising a social participation information correction unit that corrects the social participation information using the correction amount. 2. The social participation behavior analysis device according to claim 1,
the social participation information identification unit identifies the social participation information for each of a plurality of predetermined periods;
The correction amount calculation unit is a social participation behavior analysis device that calculates the correction amount for the specified period. 3. The social participation behavior analysis device according to claim 2,
The correction amount calculation unit calculates a correction amount for making social participation information for a first period correspond to social participation information for a second period among the plurality of predetermined periods. 3. The social participation behavior analysis device according to claim 2,
The correction amount calculation unit calculates a correction amount for making the social participation information for at least a first period among the plurality of specified periods correspond to a predetermined standard. 5. The social participation behavior analysis device according to claim 2,
The input unit further accepts that an intervention to support the social participation behavior has been performed,
The social participation behavior analysis device further comprises a data analysis unit that calculates an effect of the intervention based on the corrected social participation information for a specified period before and after the intervention. 6. The social participation behavior analysis device according to claim 5,
The social participation behavior analysis device further includes a risk prediction model generation unit that uses the corrected social participation information to generate a risk prediction model used to predict risks related to the user's health. 7. The social participation behavior analysis device according to claim 6,
A social participation behavior analysis device, wherein the social participation information includes at least one of the user's stay at a specific spot, employment status, and participation status in group activities. A social participation behavior analysis method for analyzing a user's social participation behavior using a social participation behavior analysis device, comprising:
An input unit receives sensor information indicating the actions of a plurality of users and environmental information indicating the environment at the time of the actions;
a social participation information identification unit that identifies social participation information indicating the social participation behavior of the user from the sensor information;
a model generation unit generates a regression model of the social participation information with respect to the environmental information for each type of social participation behavior;
A correction amount calculation unit calculates a correction amount of the social participation information corresponding to the environmental information based on the regression model;
A social participation behavior analysis method, comprising: a social participation information correction unit correcting the social participation information using the correction amount. The social participation behavior analysis method according to claim 8,
The social participation information identification unit identifies the social participation information for each of a plurality of predetermined periods;
The social participation behavior analysis method, wherein the correction amount calculation unit calculates the correction amount for the specified period. The social participation behavior analysis method according to claim 9,
The social participation behavior analysis method, in which the correction amount calculation unit calculates a correction amount for making social participation information for a first period correspond to social participation information for a second period among the plurality of specified periods. The social participation behavior analysis method according to claim 9,
The correction amount calculation unit calculates a correction amount for making the social participation information for at least a first period among the plurality of specified periods correspond to a predetermined standard. The social participation behavior analysis method according to any one of claims 9 to 11,
The input unit further accepts that an intervention to support the social participation behavior has been performed;
Furthermore, the social participation behavior analysis method includes a data analysis unit that calculates the effect of the intervention based on the corrected social participation information for each specified period before and after the intervention. The social participation behavior analysis method according to claim 12,
Furthermore, the social participation behavior analysis method includes a risk prediction model generation unit that uses the corrected social participation information to generate a risk prediction model used to predict risks related to the user's health. The social participation behavior analysis method according to claim 13,
A social participation behavior analysis method, wherein the social participation information includes at least one of the user's stay at a specific spot, employment status, and participation status in group activities.

Images