WO2021106099A1 - Action assistance information generation device, method, and program - Google Patents

Abstract

storage medium which stores information indicating the importance for each expression type, the information being defined for each assumed status of a user, with respect to a plurality of expression type candidates prepared in advance in order to deliver an action recommended to the user. In addition, a feature amount, which indicates the degree of interest in the action of the user, is acquired, the current status of the user is determined on the basis of the acquired feature amount, an expression type corresponding to the determined current status of the user is selected from among the plurality of expression type candidates on the basis of the determined current status of the user and information indicating the stored importance of the plurality of expression type candidates, and action assistance information for delivering the recommended action to the user is generated and output by using the selected expression type.

Description

Behavioral support information generators, methods and programs

One aspect of the present invention relates to a behavior support information generator, a method, and a program that generate information for supporting a behavior change of a target user.

Lifestyle-related diseases can be prevented by improving the accumulation of unhealthy lives. Therefore, various techniques have been proposed in which information on the characteristics, needs, and acceptability of the target user is investigated, and based on this information, information that supports the transformation of health behavior is provided to the target user.

For example, Non-Patent Document 1 focuses on a Transtheoretical Model (TTM) defined for the purpose of transforming an individual's behavior, and focuses on the attribute information, behavior information, environmental information, etc. of the target user. A technique is described in which the behavioral transformation stage in which the target user's situation is located is estimated based on the above, and the behavior suitable for the target user is recommended according to the estimation result.

The technique described in Non-Patent Document 1 defines the conditions for presenting recommended actions to the target user according to the situations where the actions are necessary, the actions that are likely to be executed (small change approach), and the timings at which the actions should be executed. , The recommended action is generated based on the defined presentation conditions and presented to the user. For this reason, the communication expression of the recommended behavior tends to be uniform, and it may not be possible to present the recommendation information by an appropriate communication expression according to the situation of the target user.

The present invention has been made by paying attention to the above circumstances, and one aspect is to provide a technique for presenting a recommended action by a communication expression that is easy for the user to accept according to the situation of the user.

In order to solve the above-mentioned problems, one aspect of the present invention is defined for a plurality of expression type candidates prepared in advance for transmitting a behavior recommended to the user according to the assumed situation of the user. A storage medium for storing information indicating the importance of the expression type is provided. Then, a feature amount indicating the degree of interest of the user in the action is acquired, the current situation of the user is determined based on the acquired feature amount, and the determined current situation of the user and the storage are stored. Based on the information indicating the importance of the plurality of expression type candidates, the expression type corresponding to the determined current situation of the user is selected from the plurality of expression type candidates and selected. Using the above expression type, the action support information for transmitting the recommended action to the user is generated and output.

According to the first aspect of the present invention, it is an object of the present invention to provide a technique for presenting a recommended action by a communication expression that is easy for the user to accept according to the situation of the user.

FIG. 1 is a diagram showing a configuration of a behavior change model according to an embodiment of the present invention. FIG. 2 is a diagram showing the direction of state transition in the behavior change model shown in FIG. FIG. 3 is a diagram showing a configuration of a behavior change support system according to an embodiment of the present invention. FIG. 4 is a block diagram showing a hardware configuration of a server device that operates as an action support information generation device in the system shown in FIG. FIG. 5 is a block diagram showing a software configuration of a server device that operates as an action support information generation device in the system shown in FIG. FIG. 6 is a flowchart showing an example of the procedure and processing contents of the action message generation processing by the server device shown in FIG. FIG. 7 is a flowchart showing the procedure and processing contents of the transmission expression generation process among the action message generation processes shown in FIG. FIG. 8 is a diagram showing an example of a user feature amount. FIG. 9 is a diagram showing an example of a user feature amount input to the server device shown in FIG. FIG. 10 is a diagram showing an example of situation information input to the server device shown in FIG. FIG. 11 is a diagram showing an example of the determination condition of the behavior change stage. FIG. 12 is a diagram showing an example of determination conditions for the progress phase. FIG. 13 is a diagram showing an example of information stored in the user state storage unit provided in the server device shown in FIG. FIG. 14 is a diagram showing an example of information stored in the recommended behavior information storage unit provided in the server device shown in FIG. FIG. 15 is a diagram showing an example of information stored in the expression type appearance probability storage unit provided in the server device shown in FIG. FIG. 16 is a diagram showing another example of information stored in the expression type appearance probability storage unit provided in the server device shown in FIG. FIG. 17 is a diagram showing an example of information stored in the text expression information storage unit provided in the server device shown in FIG. FIG. 18 is a diagram showing an example of information stored in the risk explanation information storage unit provided in the server device shown in FIG. FIG. 19 is a diagram showing an example of information stored in the presentation / practice situation storage unit provided in the server device shown in FIG.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings.
[One Embodiment]
(Configuration example)
(1) Behavior change model FIG. 1 shows an example of the configuration of a behavior change model used in the behavior support information generation device according to the embodiment of the present invention.
The behavior change model is composed of a behavior change stage in which the degree of the user's approach to the target behavior is defined in a plurality of states, and a progress phase in which the progress of planning and implementation for the target behavior is defined in the plurality of stages.

In this example, as a plurality of states of the behavior change stage, five stages (states) including an indifference period ST1, an interest period ST2, a preparation period ST3, an execution period ST4, and a maintenance period ST5 are defined. On the other hand, the progress phase is defined for the preparatory stage ST3 and the execution stage ST4 among the five stages of the above-mentioned behavior change stage, and is based on the three stages of the planning phase P1, the action intention phase P2, and the action phase P3. It is composed.

More specifically, the indifference period ST1 indicates a state in which there is no plan to perform a target action (preparatory action), and the interest period ST2 indicates a state in which a target action is intended. In addition, the planning phase P1 in the preparatory phase ST3 indicates the stage in which the preparatory action is planned in the near future, the action intention phase P2 indicates the stage in which the preparatory action is attempted, and the action phase P3 indicates the stage in which the preparatory action is carried out. .. Further, the planning phase P1 in the execution period ST4 indicates the stage in which the target action is planned in the near future, the action intention phase P2 indicates the stage in which the target action is attempted, and the action phase P3 indicates the stage in which the target action is executed. .. Finally, the maintenance phase ST5 indicates a state in which the target behavior is continued.

In the behavior modification model described above, each stage and phase are sequentially transitioned from the indifference period ST1 to the maintenance period ST5. FIG. 2 shows the direction of the state transition with an arrow. In Fig. 2, (1) indicates the possibility of transition to the stage, (2) indicates the possibility of increasing from the planning phase to the action will phase, and (3) indicates the possibility of transition from the action intention phase to the action phase. It is a thing.

(2) System FIG. 3 is a diagram showing an overall configuration of a behavior change support system according to an embodiment of the present invention. This behavior change support system is provided via a network NW between a server device SV that operates as a behavior support information generation device according to an embodiment of the present invention and user terminals MT1 to MTn that are used by a plurality of users. It enables data communication.

User terminals MT1 to MTn consist of smartphones, tablet terminals, wearable terminals, personal computers, etc., and are used not only by the target users but also by supporters such as family members, trainers, and medical personnel who support the behavior change of the target users. May be good.

The network NW is composed of, for example, an IP (Internet Protocol) network such as the Internet and an access network for accessing this IP network. As the access network, for example, a wired and wireless public network and a wired and wireless LAN (Local Area Network) are used.

(3) Server device SV
4 and 5 are block diagrams showing the hardware configuration and software configuration of the server device SV, respectively.
The server device SV is composed of, for example, a cloud server or a Web server. The server device SV includes a control unit 1 having a hardware processor such as a central processing unit (CPU), and a storage unit 2 and a communication interface (communication I / F) 3 are provided to the control unit 1. It is connected via the bus 4.

The communication I / F3 is provided with an interface corresponding to, for example, a public data network, and data is transmitted between the user terminals MT1 to MTn via the network NW. The communication I / F3 may be provided with other interfaces such as a wired LAN depending on the installation location and the operator.

The storage unit 2 is a combination of, for example, a non-volatile memory such as an HDD (Hard Disk Drive) or a Solid State Drive (SSD) that can be written and read at any time, and a ROM (Read Only Memory) and a RAM (Random Access Memory). It is provided with a storage medium composed of the above, and has a program storage area and a data storage area. The configuration of the storage medium is not limited to the above configuration. In the program storage area, in addition to middleware such as an OS (Operating System), programs necessary for executing various control processes according to an embodiment of the present invention are stored.

The data storage area includes a stage determination condition storage unit 21, a phase determination condition storage unit 22, a user state storage unit 23, a recommended action information storage unit 24, an expression type appearance probability storage unit 25, and a sentence expression information storage. A unit 26, a risk explanation information storage unit 27, and a presentation / practice situation storage unit 28 are provided.

The stage determination condition storage unit 21 stores the stage determination conditions for determining which stage the target user's state corresponds to in association with each of the five stages of the behavior change model illustrated in FIG. ..

The phase determination condition storage unit 22 stores the phase determination conditions for determining which phase the target user's stage corresponds to in association with each of the three stages of the behavior change model illustrated in FIG. There is.

The user state storage unit 23 stores the determination result of the behavior change stage and phase of the user and the health risk level of the user in association with the user's identification information (user ID).

The recommended behavior information storage unit 24 stores content representing the behavior content recommended to the user and the category of the content (for example, "exercise", "meal").

The expression type appearance probability storage unit 25 has a behavior change stage (“indifferent period”, “interest period”) for each expression type (“information provision”, “proposal”, “declaration guidance”, “instruction”, “command”). , "Preparation period", "Execution period", "Maintenance period") It has multiple tables that store the appearance probabilities for each. The appearance probability is used as information indicating the importance of the expression type. User attributes (for example, gender and age) may be defined in each table.

In the sentence expression information storage unit 26, a plurality of sentence expression examples are provided with identification information (ID) for each expression type (“information provision”, “proposal”, “declaration guidance”, “instruction”, “command”). It is stored in the associated state.

The risk explanation information storage unit 27 stores an example of content representing the content of the risk.

The presentation / practice situation storage unit 28 displays the presentation history of the behavior support message to the user and the information representing the user's practice result for the presented behavior support message in the user's behavior change stage and attributes (for example, gender and age). Remember with.

The control unit 1 includes a stage determination unit 11, a phase determination unit 12, a health risk information acquisition unit 13, a recommended action selection unit 14, and a transmission expression as various control functions for realizing one embodiment of the present invention. It has a generation unit 15, an appearance probability update unit 16, and a reaction log acquisition unit 17. All of these control functions are realized by causing the hardware processor of the control unit 1 to execute the program stored in the program storage area of the storage unit 2.

The stage determination unit 11 acquires the user's feature amount (which may include situation information), and determines the user's behavior change stage based on the acquired user feature amount. Then, the determination result of the behavior change stage is stored in the user state storage unit 23 in association with the user ID.

The user feature amount is information that reflects the degree of interest of the user in the target behavior, and acquisition of the user feature amount is, for example, transmitting a questionnaire-type question list to the user terminals MT1 to MTn of the target user. Then, the answer information to the above question list is received from the user terminals MT1 to MTn. Further, in the determination of the behavior change stage, for example, the acquired user feature amount is collated with a plurality of stage determination conditions stored in the stage determination condition storage unit 21, and the current state of the target user is determined. It is performed by determining which stage corresponds to.

When the stage determination unit 11 determines that the current state of the target user corresponds to the "preparation period" or the "execution period", the phase determination unit 12 determines the user feature amount as the phase determination condition. By collating with a plurality of phase determination conditions stored in the storage unit 22, it is determined which phase the current state of the target user corresponds to. Then, the determination result of the above phase is stored in the user state storage unit 23 in association with the user ID.

The health risk information acquisition unit 13 receives the information representing the user's health risk sent from the user terminals MT1 to MTn, associates the received information representing the health risk with the user ID, and stores it in the user state storage unit 23. Let me. The information representing the health risk is expressed as, for example, the degree (risk level) of the health risk determined from the result of the health diagnosis.

The recommended action selection unit 14 operates, for example, triggered by the end of the stage and the determination event of the phase, and selectively reads the recommended action content representing the content of advice to the user from the recommended action information storage unit 24. Then, the read recommended action content is given to the transmission expression generation unit 15.

The transmission expression generation unit 15 performs the following processing.
(1) For each user, the behavior change stage of the user and its phase are read from the user state storage unit 23, and the state and stage of the user are determined based on the read behavior change stage and the phase. As a result of this determination, the process shifts to the action support message generation process when the user's state is the indifference period or the interest period, and when the user's state is the preparation period or the action period and the planning phase.

(2) Appearance probability for each expression type from the above-mentioned expression type appearance probability storage unit 25 according to the user's current behavior change stage and its phase determined earlier in the action support message generation process, and the health risk level. Is read to generate a conditional expression, and the expression type is selected by the weighted random sampling method.

(3) From the sentence expression information storage unit 26, select a sentence expression corresponding to the selected expression type, and send an action support message based on this sentence expression and the content selected by the recommended action selection unit 14. Generate. Then, the generated action support message is transmitted to the corresponding user terminals MT1 to MTn, and information representing the presentation history of the action support message is written in the presentation / practice situation storage unit 28.

When the reaction log is returned from the user terminals MT1 to MTn to which the action support message is sent, the reaction log acquisition unit 17 receives and presents the reaction log, and the corresponding presentation of the practice status storage unit 28. Write the practice flag "1" in the history record.

When the practice flag "1" is written in the presentation / practice situation storage unit 28, the appearance probability update unit 16 is, for example, machine learning based on the presentation history record stored in the presentation / practice situation storage unit 28. To update the appearance probability.

(Operation example)
Next, the operation of the server device SV configured as described above will be described.
6 and 7 are flowcharts showing the processing procedure and processing contents of the action support message generation control by the server device SV.

(1) Acquisition of User Feature Amount Under the control of the stage determination unit 11, the server device SV first executes a process of acquiring the feature amount for each target user in step S11 as follows.
That is, the stage determination unit 11 first transmits a questionnaire-style question list to the user terminals MT1 to MTn of the target user at each preset acquisition timing of the day, for example. In this operation, for example, an e-mail including the URL (Uniform Resource Locator) of the server device SV is transmitted to the user terminals MT1 to MTn, and the user accesses the server device SV based on the above URL and downloads the question list. It is done by doing.

The question list is selected from, for example, a plurality of lists stored in the storage unit 2. If the attribute information of the target user has been acquired in advance, the list corresponding to this attribute information may be selected. In this way, it is possible to select and send an appropriate question list according to the age, gender, occupation, medical history, etc. of each target user.

On the user terminals MT1 to MTn, the user inputs an answer while looking at the displayed question list. Then, when the user performs the end operation after the input of the answer is completed, the content of the input answer, the input date and time, and the answer information including the user ID or the user terminal ID are transmitted from the user terminals MT1 to MTn to the server device SV. Will be done.

On the other hand, when the answer information is transmitted from the user terminals MT1 to MTn, the server device SV receives the answer information via the communication I / F3 under the control of the stage determination unit 11. Then, the received response information is stored in the temporary storage area in the storage unit 2 as a user feature amount.

FIG. 8 shows an example of the types of user features and situation information, their collection methods, and collection timings. FIG. 9 shows a question list for acquiring the user feature amount and an example of the answer, and FIG. 10 shows an example of the answer of the situation information. The items in the question list can be set arbitrarily, and the answers to the questions may be omitted.

(2) Judgment of behavior change stage When the new user feature amount is acquired, the server device SV uses the acquired user feature amount in step S12 as a stage determination condition under the control of the stage determination unit 11. By collating with the stage determination condition stored in the storage unit 21, it is determined which stage the current state of the target user corresponds to.

For example, it is assumed that a plurality of stage determination conditions shown in FIG. 11 are currently stored in the stage determination condition storage unit 21. In this state, for example, assuming that the user feature amount shown in FIG. 9 is acquired, the stage determination unit 11 indicates that the content of the user feature amount is "target execution plan: yes" and "are you thinking of acting now: yes". Therefore, it is determined that the current state of the target user is the "execution period".

The determination result of the state (stage) of the target user obtained as described above is stored in the user state storage unit 23 in a state associated with the user ID under the control of the stage determination unit 11.

(3) In-stage phase determination The server device SV determines the phase when it is determined that the current state of the target user corresponds to the "preparation period" or "execution period" in the determination of the behavior change stage. Under the control of unit 12, in step S13, the user feature amount is collated with a plurality of phase determination conditions stored in the phase determination condition storage unit 22, and the current situation of the user corresponds to which phase in the stage. Determine if you want to.

For example, it is assumed that the phase determination condition storage unit 22 now stores the phase determination condition shown in FIG. In this state, for example, assuming that the user feature amount shown in FIG. 9 is acquired, the phase determination unit 12 includes "action implementation plan: yes" and "are you planning to act now: yes" included in the user feature amount. , "Are you acting now: No" is collated with the phase determination condition stored in the phase determination condition storage unit 22. As a result, the current stage of the user is determined to be the "action will phase".

The determination result of the stage (phase) of the target user obtained as described above is stored in the user state storage unit 23 in a state associated with the user ID under the control of the phase determination unit 12.

FIG. 13 shows an example of the behavior change stage and the phase stored in the user state storage unit 23. When determining the behavior change stage and its phase, situation information may be used in addition to the user's feature amount. FIG. 10 shows an example of the acquired situation information. In this example, the situation information includes “slightly tired” and “outside air temperature 9 ° C.”.

(4) Acquisition of health risk information The server device SV monitors, under the control of the health risk information acquisition unit 13, whether or not the acquisition date and time set in advance for each user has been reached in step S14. When the acquisition date and time arrives in this state, a health risk information acquisition request is transmitted to the corresponding user terminals MT1 to MTn. Then, in response to this request, the health risk information transmitted from the user terminals MT1 to MTn is received via the communication I / F3, and the received health risk information is associated with the user ID and stored in the user state storage unit 23. ..

As the health risk information, for example, the user judges the risk level related to the health condition by himself / herself based on the result of the health diagnosis in three stages of "low", "medium", and "high". However, in addition to that, for example, the health risk information acquisition unit 13 acquires the user's health examination data from the user terminals MT1 to MTn, determines the health risk level based on the acquired health examination data, and determines the determination. The result may be stored in the user state storage unit 23 as health risk information.

In addition, in the acquisition of the health risk information, the health risk information acquisition unit 13 monitors the transmission request transmitted from each user terminal MT1 to MTn at an arbitrary timing, and receives the health risk information in response to the transmission request. It may be done by doing. It should be noted that the health risk information is not indispensable for generating the action support message, and may be acquired only when the user can provide it.

(5) Selection of recommended behavior When the above stage and phase determination event or health risk information acquisition event is completed, the server device SV subsequently gives advice to the user in step S15 under the control of the recommended behavior selection unit 14. The recommended action content representing the above is selected from the recommended action information storage unit 24 and read out. Then, the recommended action selection unit 14 gives the content of the selected recommended action to the transmission expression generation unit 15.

The recommended action content may be randomly selected from, for example, a plurality of contents stored in advance, or attribute information such as the user's gender, age, and hobbies may be acquired in advance, and this user attribute may be obtained. Content suitable for the user may be selected based on the information.

FIG. 14 shows an example of the content stored in the recommended behavior information storage unit 24. In this example, if the user is a commuter, "walking for 5 minutes while commuting" is selected, and if the user is not commuting but is interested in exercising, "5 squats" is selected. To commute. For users who are interested in eating, "eat vegetables first" is selected.

(6) Generation of a transmission expression of the action support message When the selection process of the recommended action content by the recommended action selection unit 14 is completed, the server device SV then performs an action in step S16 under the control of the transmission expression generation unit 15. The process of generating the transmission expression of the support message and transmitting the transmission support message is executed as follows.

FIG. 7 is a flowchart showing a processing procedure and processing contents by the transmission expression generation unit 15. The transmission expression generation unit 15 first determines whether or not it is appropriate to provide behavior support to the target user based on the current behavior change stage of the target user and the phase thereof. For example, the transmission expression generation unit 15 first reads the behavior change stage of the target user and its phase from the user state storage unit 23 in step S161. Then, in step S162, it is determined whether or not the state of the target user is the "preparation period" or the "behavioral period" based on the read behavior change stage and its phase. As a result of this determination, if the state of the user is "indifferent period" or "interest period", it is determined that the user is the target person of the action support, and the process shifts to the action support message generation process.

On the other hand, if the user's state is the "preparation period" or the "action period", the transmission expression generation unit 15 determines in step S163 whether or not the stage in the above stage of the user is the planning phase. Then, in the planning phase, the target user is determined to be the target person of the action support in this case as well, and the process shifts to the action support message generation process.

When the process shifts to the action support message generation process, the transmission expression generation unit 15 generates a conditional expression based on the appearance probability stored in the expression type appearance probability storage unit 25 in step S164. Equation (1) shows an example of the generated conditional expression. In equation (1), the expression type k of Pk is k = 1: "information provision", k = 2: "proposal", k = 3: "declaration guidance", k = 4: "instruction", k = 5: Indicates the appearance probability at the time of "command".

Next, the transmission expression generation unit 15 selects an expression type based on the generated conditional expression (1) in step S165.
For example, it is assumed that the expression type appearance probability storage unit 25 now stores the appearance probability data as shown in FIG. Here, 25a, 25b, and 25c show data tables when the health risk levels are "low", "medium", and "high", respectively. In this state, assuming that the current state of the target user is the behavior change stage: "execution period", the in-stage phase: "planning phase", and the risk level: "low", the conditional expression (1) is as follows. Become.

Then, for example, if random (0 to 100) = 75, the transmission expression generation unit 15 selects k = 2, that is, "proposal" as the expression type k.

Note that the appearance probability data stored in the expression type appearance probability storage unit 25 may be defined for each user attribute. FIG. 16 shows an example of the appearance probability data in this case. In FIG. 16, as in FIG. 15, 25a, 25b, and 25c show data tables when the health risk levels are “low”, “medium”, and “high”, respectively.

In this case, the transmission expression generation unit 15 acquires user attribute information from the user terminals MT1 to MTn. Then, in consideration of the gender and age included in the acquired user attribute information, the corresponding expression type appearance probability data is read from the expression type appearance probability storage unit 25, and the above conditional expression (1) is generated by the user. Select the behavior change stage and its phase, and the expression type corresponding to the gender and age.

Subsequently, in step S166, the transmission expression generation unit 15 selectively reads out the sentence expression information corresponding to the selected expression type from the sentence expression information storage unit 26. Then, an action support message for the user is generated based on the read sentence expression information and the recommended action content previously selected by the recommended action selection unit 14.

For example, it is assumed that the sentence expression information storage unit 26 now stores candidates for the sentence expression information shown in FIG. 17, and as described above, "proposal" is selected as the expression type in step S165. In this case, the transmission expression generation unit 15 reads, for example, "Would you like to take [recommended action]?" From a plurality of candidates for sentence expression information corresponding to the "proposal". Then, by synthesizing the "squat 5 times" previously selected by the recommended action selection unit 14 at the position of [recommended action] in this sentence expression information,
"Why don't you squat 5 times?"
Generate an action support message that includes the content of the presentation.

Also, for example, "command" is selected as the expression type, and from among multiple candidates for sentence expression information corresponding to this "command", for example, "There is [risk]. [Recommended action] must be done!" Is selected. In this case, the transmission expression generation unit 15 reads the risk content from the risk explanation information storage unit 27.

For example, assuming that the risk explanation information storage unit 27 stores the content for risk explanation shown in FIG. 18, the transmission expression generation unit 15 reads out the “risk of amputation or artificial dialysis in the case of diabetes”. Then, the transmission expression generation unit 15 synthesizes the "risk of leg amputation or artificial dialysis when diabetes occurs" read from the risk explanation information storage unit 27 with the [risk] of the sentence expression information, and further, the above sentence. By synthesizing the "squat 5 times" previously selected by the recommended action selection unit 14 with the [recommended action] of the expression information,
"If you have diabetes, you run the risk of amputation and dialysis. You have to squat 5 times!"
Generate a message containing the content of the presentation.

Finally, in step S167, the transmission expression generation unit 15 transmits the generated action support message from the communication I / F3 to the corresponding user terminals MT1 to MTn. As the transmission means, for example, an e-mail or an SNS message is used. The transmitted action support message is received by the user terminals MT1 to MTn, and the presented content is displayed on, for example, a display.

Further, the transmission expression generation unit 15 writes information representing the presentation history in the presentation / practice situation storage unit 28 based on the transmitted action support message. In the presentation / practice status storage unit 28, for example, as shown in FIG. 19, a "presentation date / time", a "practice date / time", a "presentation flag" indicating the presence / absence of a presentation action, and a "presentation flag" indicating the presence / absence of a user's practice (user action) are shown. The "practice flag", "user ID", "expression type" of the sent action support message, "behavior change stage" of the user, and "user attribute (gender and age)" are stored. Of these pieces of information, the transmission expression generation unit 15 represents the presentation history, such as "presentation date and time", "presentation flag", "user ID", "expression type", "behavior change stage", and "user attribute (gender)". And age) ”, write the information respectively.

(7) Update of Expression Type Appearance Probability Each time the server device SV transmits an action support message, the expression type appearance probability storage unit 25 stores each expression type according to the user's reaction to the action support message. Performs processing to update the appearance probability of the presented action.

Upon receiving the above action support message, the user determines whether or not the content of the received message is likely to be useful to him / her. If it seems to be useful, it will act according to the content presented, while if it is judged that it is unlikely to be useful, it will not act. The user logs the presence or absence of the practice of this action and transmits it from the user terminals MT1 to MTn to the server device SV.

When the reaction log is returned from the user terminals MT1 to MTn, the server device SV receives the reaction log in step S17 under the control of the reaction log acquisition unit 17. Then, the reaction log acquisition unit 17 stores the "practice date and time" in the presentation / practice status storage unit 28 based on the received reaction log, and sets the "practice flag" to "1" or "0". .. If the reaction log is not returned within a predetermined time after the action support message is sent, it is considered that the user has not responded to the presented content, and the "practice flag" is set to "0".

When the practice flag "1" is set in the presentation / practice situation storage unit 28, the appearance probability update unit 16 determines the behavioral transformation stage and user attributes (gender and age) stored in the presentation / practice situation storage unit 28. For each combination of, using the information of "presentation flag", "practice flag", "expression type" and "behavior transformation stage", the presentation action of each expression type by machine learning such as SVM (Support Vector Machine) Update the appearance probability. At that time, in machine learning, learning is performed so as to satisfy the update formula (2) shown below.

However, S_type is the presentation action for each expression type (type is the expression type), S_all is the set of presentation actions, P (S_type) is the appearance probability of the presentation action for each expression type before update, and P_new (S_type) is after the update. The appearance probability of the presented action for each expression type, A represents the combination of the behavioral transformation stage and the user attribute, and D | S_type represents the user's practice status (user action) for the presented content.

For example, when updating the appearance probability corresponding to the "indifference period" of a man in his thirties shown in FIG. 16, the appearance probability updating unit 16 learns so as to satisfy the following equation.

As a result, for example, a new appearance probability of the presented action for each expression type shown in the following formula can be obtained.

The appearance probability update unit 16 updates the value of the corresponding appearance probability stored in the expression type appearance probability storage unit 25 to the new appearance probability value.

In the above update formula (2), only the behavior change stage may be used without considering the user attributes (gender and age). Further, instead of the practice flag, information indicating the user's willingness to consent (for example, I think it is valuable or I want to try it) used in SNS or the like may be used.

(Action / effect)
As described above, in one embodiment, the user's current behavior change stage and its phase are determined based on the acquired user feature amount, and the user is determined based on the determined behavior change stage and its phase. Determine if action support is needed. Then, when it is determined that behavior support is necessary, the appearance probability for each expression type corresponding to the user's current behavior change stage and its phase, and the user's health risk level is read from the expression type appearance probability storage unit 25. The conditional expression (1) is generated, and the expression type corresponding to the current state of the user is selected using this conditional expression (1). Then, an action support message is generated by synthesizing the recommended action for the user and the content of the risk explanation with the sentence expression corresponding to the selected expression type, and is transmitted to the corresponding user terminal.

Therefore, since the action support message is presented to the user only to the user who needs the action support, it is possible to provide effective action support according to the user's situation. In addition, when generating a behavior support message, a transmission expression that satisfies a predetermined condition is selected and selected based on the user's current behavior change stage and its phase, and the appearance probability of the expression type corresponding to the health risk level. An action support message using the transmitted expression is generated and presented to the user. Therefore, it is possible to generate and present an action support message that is easy for the user to accept in response to the user's situation, which effectively motivates the user to take action and increases the practice rate for recommended actions. Become.

Further, in one embodiment, the reaction log of the user to the presentation of the action support message is acquired, and is stored in the expression type appearance probability storage unit 25 in consideration of the practice status of the recommended action represented by the acquired reaction log. The probability of appearance of the expression type corresponding to the corresponding behavioral transformation stage and health risk level is updated. Therefore, as the number of behavior support processes increases, it becomes possible to select a communication expression that is more suitable for the user's situation.

[Other Embodiments]
The present invention is not limited to the above embodiment. For example, in one embodiment, the case where the function of the action support information generator is provided in the server device SV such as a Web server or a cloud server has been described as an example, but a smartphone or a smartphone in which the user uses the function of the action support information generator. It may be provided in a user terminal such as a tablet terminal or a personal computer.

In addition, the types of behavior modification targeted by the present invention include not only lifestyle-related disease prevention and lifestyle improvement for dieting, but also exercise habits, dietary habits, learning habits, and behaviors for quitting quitting. And maintenance etc. are also conceivable.

In addition, the type of user feature amount used to judge the user's situation, the judgment method of the behavior change stage and its phase, the conditional expression when selecting the expression type of the communication expression, the type and content of the recommended behavior information content, and the text. The type and content of the expression information, the type and content of the content for explaining the risk, the procedure and the processing content of the action support message generation process, and the like can be variously modified without departing from the gist of the present invention.

In short, the present invention is not limited to the above-described embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, some components may be removed from all the components shown in the embodiments. In addition, components from different embodiments may be combined as appropriate.

SV ... Server device MT1 to MTn ... User terminal NW ... Network 1 ... Control unit 2 ... Storage unit 3 ... Communication I / F
4 ... Bus 11 ... Stage judgment unit 12 ... Phase judgment unit 13 ... Health risk information acquisition unit 14 ... Recommended action selection unit 15 ... Transmission expression generation unit 16 ... Appearance probability update unit 17 ... Reaction log acquisition unit 21 ... Stage judgment condition memory Part 22 ... Phase judgment condition storage unit 23 ... User state storage unit 24 ... Recommended behavior information storage unit 25 ... Expression type appearance probability storage unit 26 ... Sentence expression information storage unit 27 ... Risk explanation information storage unit 28 ... Presentation / practice situation memory Department

Claims (8)

1. A storage medium that stores information indicating the importance of the expression type, which is defined according to the situation assumed by the user, for a plurality of expression type candidates prepared in advance to convey the recommended action to the user. ,
   A situation determination unit that acquires a feature amount indicating the degree of interest of the user in the action and determines the current situation of the user based on the acquired feature amount.
   Based on the determined current situation of the user and the stored information indicating the importance of the candidates of the plurality of expression types, the plurality of expression types corresponding to the determined current situation of the user are selected. The expression type selection unit that selects from the expression type candidates, and
   An action support information generation device including a generation unit that generates and outputs action support information for transmitting the recommended action to the user using the selected expression type.
2. The user's situation is based on a behavior change stage that defines the degree of the user's efforts on the behavior in a plurality of states, and a progress phase that defines the progress of the user's planning and implementation on the behavior in a plurality of stages. The behavior support information generation device according to claim 1, which is represented.
3. The expression type selection unit determines whether or not the determined current situation of the user is a situation requiring the presentation of the action support information, and is a situation requiring the presentation of the action support information. The action support information generation device according to claim 1, wherein when it is determined, the process of selecting the expression type is performed.
4. The expression type selection unit is generated by reading the appearance probability of the expression type candidate corresponding to the determined current situation of the user from the storage medium and generating a conditional expression for selecting the expression type. The action support information generation device according to claim 1, wherein an expression type corresponding to the current situation of the user is selected by using a conditional expression.
5. A risk information acquisition unit for acquiring information representing risks in the user's life is further provided.
   The expression type selection unit determines based on the current situation of the determined user, the acquired information indicating the risk, and the stored information indicating the importance of the plurality of expression type candidates. The action support information generation device according to claim 1, wherein an expression type corresponding to the current situation and risk of the user is selected from the plurality of expression type candidates.
6. A reaction information acquisition unit that acquires information representing the user's reaction to the generated action support information, and a reaction information acquisition unit.
   Based on the acquired information representing the reaction and the information representing the current situation of the user used to generate the action support information, the information representing the importance of the expression type stored in the storage medium is provided. The action support information generation device according to claim 1, further comprising an update unit for updating.
7. It is a behavior support information generation method executed by an information processing device having a processor.
   For a plurality of expression type candidates prepared in advance to convey the recommended action to the user, information indicating the importance of the expression type defined according to the expected situation of the user is stored in the storage medium. The process and
   A process of acquiring a feature amount representing the degree of interest of the user in the behavior and determining the current situation of the user based on the acquired feature amount.
   Based on the determined current situation of the user and the stored information indicating the importance of the candidates of the plurality of expression types, the plurality of expression types corresponding to the determined current situation of the user are selected. The process of selecting from the candidates for the expression type and
   A behavior support information generation method including a process of generating and outputting behavior support information for transmitting the recommended behavior to the user using the selected expression type.
8. A program for causing a processor included in the action support information generator to execute the processing of each part included in the action support information generator according to any one of claims 1 to 6.

Images