Action suggestion device, action suggestion method, and program

Abstract

To perform an action proposal in consideration of a relationship between a proposed person and a proposer.SOLUTION: An action suggestion device according to an embodiment includes an index information acquisition unit, a relationship information acquisition unit, a suggestion acquisition unit, and an information output unit. The index information acquisition unit acquires index information representing an action index of a proposed person. The relationship information acquisition unit acquires relationship information indicating a relationship between the proposed person and a proposer. The proposal acquisition unit acquires an action proposal for the proposed person on the basis of the index information and the relationship information. The information output unit outputs information based on the action proposal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The embodiments disclosed in this specification and the drawings relate to an action suggestion device, an action suggestion method, and a program. [Background technology]

[0002] In order to improve the behavior of users with illnesses such as lifestyle-related diseases, suggestors such as doctors or nurses may make behavioral suggestions and encourage the users to change their behavior. To prevent frailty in users, local government-sponsored participatory programs may also be used to encourage behavioral changes by suggesting behaviors from people around the user. However, because users are real people with different lifestyles and differing relationships with suggestors, some users may find it difficult to faithfully adhere to the suggested behavioral changes.

[0003] On the other hand, there is a known behavior change support device that encourages behavior change based on the user's actual situation. This type of behavior change support device gradually changes the suggestor based on the user's behavioral results after the suggestor has encouraged the user to change their behavior. With this behavior change support device, it is possible to select an appropriate level of suggestor to encourage behavior change, for example, depending on whether the user is actively improving their behavior.

[0004] Although the behavior change support device described above usually presents no particular problems, the inventor's research has revealed that there is room for improvement in that the behavioral suggestions from the proposer do not take into account the relationship between the user and the proposer. For example, if the relationship between the user and the proposer is low, the user may not accept the behavioral suggestions, and their behavior may not improve. Furthermore, if the behavior does not improve and the proposer is changed repeatedly, there is a concern that this may lead to a decrease in the user's motivation. Therefore, the inventor's research has revealed that it is desirable to make behavioral suggestions that take into account the relationship between the proposer and the user who is the recipient of the suggestions. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-13208 Summary of the Invention [Problem to be solved by the invention]

[0006] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to propose an action that takes into account the relationship between the proposer and the person making the proposal. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0007] An action suggestion device according to an embodiment includes an index information acquisition unit, a relationship information acquisition unit, a proposal acquisition unit, and an information output unit. The index information acquisition unit acquires index information representing a behavioral index of a proposer. The relationship information acquisition unit acquires relationship information representing a relationship between the proposer and the proposer. The proposal acquisition unit acquires an action suggestion for the proposer based on the index information and the relationship information. The information output unit outputs information based on the action suggestion. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing a system including an action suggestion device according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing the configuration of the information storage device of FIG. [Figure 3] FIG. 3 is a schematic diagram illustrating an example of relationship information according to an embodiment. [Figure 4] FIG. 4 is a schematic diagram illustrating an example of proposed selection information according to an embodiment. [Figure 5] FIG. 5 is a block diagram showing the configuration of the action suggestion device of FIG. [Figure 6]FIG. 6 is a sequence diagram for explaining the operation in one embodiment. [Figure 7] FIG. 7 is a sequence diagram for explaining the operation in one embodiment. [Figure 8] FIG. 8 is a schematic diagram illustrating proposed selection information according to a first modified example of an embodiment. [Figure 9] FIG. 9 is a schematic diagram illustrating proposed selection information according to a second modified example of an embodiment. [Figure 10] FIG. 10 is a block diagram illustrating a configuration of an action suggestion device according to a third modified example of an embodiment. [Figure 11] FIG. 11 is a schematic diagram for explaining a learning control function according to a third modified example of the embodiment. [Figure 12] FIG. 12 is a schematic diagram illustrating a trained model according to a third modified example of an embodiment. [Figure 13] FIG. 13 is a schematic diagram illustrating a trained model according to a third modified example of an embodiment. [Figure 14] FIG. 14 is a block diagram showing a system including an information storage device including an action suggestion device according to a fourth modified example of an embodiment. [Figure 15] FIG. 15 is a block diagram showing the configuration of the information storage device of FIG. [Figure 16] FIG. 16 is a sequence diagram illustrating an operation in the fourth modified example of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] An action suggestion device, an action suggestion method, and a program according to an embodiment will be described below with reference to the drawings. In the following description, components having substantially the same functions and configurations are denoted by the same reference numerals, and redundant descriptions will be provided as appropriate and necessary.

[0010] FIG. 1 is a block diagram showing a system including an action suggestion device according to one embodiment. The system includes an information storage device 10, a medical AI (artificial intelligence) system 20, a user terminal 30, and an action suggestion device 40. The information storage device 10, the medical AI system 20, the user terminal 30, and the action suggestion device 40 are connected to each other via a network, whether wireless or wired, so that they can communicate with each other. The network is, for example, a local area network (LAN). However, the connection line is not limited to a LAN as long as security is ensured by a virtual private network (VPN) or the like. In this case, the network may be, for example, a public communication line such as the Internet.

[0011] The information storage device 10 is a database system that stores and manages relationship information and proposed selection information, which will be described later. The information storage device 10 is realized, for example, by a general server device or a cloud server. As shown in FIG. 2, the information storage device 10 includes, for example, a memory 11, a communication interface 12, and a processing circuit 13. The memory 11, the communication interface 12, and the processing circuit 13 are connected to each other so as to be able to communicate with each other, for example, via a bus.

[0012] The memory 11 is a storage device such as a hard disk drive (HDD), a solid state drive (SSD), or an integrated circuit storage device that stores various information. The memory 11 may also be a drive that reads and writes various information from and to a portable storage medium such as a CD-ROM drive, a DVD drive, or a flash memory. The memory 11 stores, for example, a program that causes the processing circuit 13 to implement a management function 131, such as creating, saving, searching, and outputting relationship information T1 and proposal selection information T2, as shown in FIGS. 3 and 4. The program may be stored in a non-transitory computer-readable storage medium, distributed, read from the storage medium, and installed in the memory 11.

[0013] Here, the relationship information T1 is personal information about a specific user and, as shown in FIG. 3, is information that represents the relationship between the proposer and the proposer. For example, the relationship information T1 may include at least one of the intimacy and the relationship between the proposer and the proposer as the relationship. Furthermore, the relationship information T1 may also include supplemental information, including at least one of the residential relationship, hobbies, and habits between the proposer and the proposer as the relationship. In FIG. 3, the user IDs (A, B, C, D, E, . . .) in each row of the left column identify the proposer, and the user IDs (A, B, C, D, E, . . .) in each column except the left column identify the proposer. The relationship between the proposer and the proposer is described at the intersection of the row of the user ID of the proposer and the column of the user ID of the proposer. In the example of FIG. 3, the relationship includes both the intimacy and the relationship, and may further include supplemental information if desired. The relationship may also be a combination of two or more of the intimacy, the relationship, and the supplemental information.

[0014] Intimacy is the degree to which the proposer has a favorable impression and trust for the proposer, and the higher the intimacy, the more likely the proposer is to agree with the proposal from the proposer. Multiple levels of intimacy are available, such as three levels: high, medium, and low. The term "intimacy" may be changed to other terms that have the same meaning as the aforementioned "intimacy," such as "trust" or "agreement."

[0015] The relationship is an attribute that indicates the human relationship between the proposer and the proposer, and can be, for example, family (couple, parent and child, siblings, grandparents, grandchildren, etc.), relatives (aunts and uncles, etc.), friends, acquaintances, workplace (bosses, subordinates, etc.), etc. Each attribute can be expanded as appropriate, for example, as shown in Figure 3, to indicate a relationship between a "couple" as "husband" or "wife" from one perspective.

[0016] The supplemental information is information that supplements at least one of the intimacy and the relationship, and can be, for example, residential relationship (living together, nearby, distant), hobbies and interests (e.g., golf, watching golf), habits, pets, etc., as appropriate. Note that the supplemental information is not limited to one piece, and may include two or more pieces, such as "neighborhood" and "hobby: golf." For example, if the relationship is "living together" or "neighborhood" and the hobbies, interests, and habits are common, it is possible to suggest activities to do together.

[0017] The proposal selection information T2 is general information about an unspecified user and is information for selecting an action proposal based on behavioral indicators and relationships. Specifically, as shown in Fig. 4, the proposal selection information T2 is information that associates index information representing the behavioral indicators of the proposal recipient, relationship information representing the relationship between the proposal recipient and the proposer, and an action proposal list including each action proposal for the proposal recipient. The proposal selection information T2 makes it possible to select any action proposal in the action proposal list based on the index information and relationship information.

[0018] The index information represents, for example, a behavioral index such as "It is good to exercise about XX times a day." Note that the behavioral index is not limited to exercise, and any other lifestyle habits other than exercise, such as "Limit your diet to XX" or "Quit smoking," or any index related to hospital life, such as "Do not go out without permission" or "Avoid trouble with other patients," can be used as appropriate.

[0019] As described above, the relationship information includes at least one of the intimacy level, the relationship, and the supplementary information.

[0020] The behavioral suggestion list includes behavioral suggestions that are easy to implement depending on the relationship with the suggestee, from the perspective of helping the suggestee achieve the behavioral indicators. The behavioral suggestions are information suggesting behaviors such as, for example, cycling (accompanied), walking (with a dog), jogging (accompanied), walking (accompanied), swimming, jogging, and light exercise. The term "accompanied" in the behavioral suggestions refers to a state in which the suggestee and the proposer are face-to-face or able to converse, and does not necessarily require the proposer to be in close proximity to the suggestee. For example, jogging (accompanied) may refer to an action in which the proposer, who is remotely connected to the proposer via a video call via user terminal 30, jogs while holding the user terminal 30. Similarly, "accompanied" does not necessarily require the proposer to perform the same behavior as the suggestee. For example, jogging (accompanied) may refer to the suggestee jogging while the proposer accompanies the suggestee on a bicycle. Such "accompanied" behavior corresponds, for example, to a high or medium level of intimacy.

[0021] The communication interface 12 performs data communication between the medical AI system 20, the user terminal 30, and the action suggestion device 40, which are connected via a network.

[0022] The processing circuitry 13 is a processor that functions as the core of the information storage device 10. The processing circuitry 13 executes a program stored in the memory 11 to realize a management function 131 corresponding to the program.

[0023] The management function 131 manages the relationship information T1 and the proposal selection information T2 in the memory 11. For example, the management function 131 communicates with the user terminal 30 and the action suggestion device 40. In addition, the management function 131 executes a read / write process for the relationship information T1 and a read process for the proposal selection information T2 in association with the communication. In addition, the management function 131 communicates with an administrator terminal (not shown) and stores the proposal selection information T2 in the memory 11 in advance.

[0024] The medical AI system 20 is a system in which a medical AI generates behavioral indicators for a proposed user based on status information indicating the user's status transmitted from a user terminal 30 and transmits the generated behavioral indicators to an action suggestion device 40. The medical AI is a trained model that has been machine-learned using training data in which status information is input data and behavioral indicators are output data. The medical AI system 20 may be implemented, for example, as a cloud server equipped with memory, a communication interface, a processing circuit, and the like (not shown). The cloud server is a server built in a cloud environment, and the processing circuit executes a program stored in memory to receive status information, generate behavioral indicators from the status information using the medical AI stored in the memory, and transmit the generated behavioral indicators. A cloud environment refers to an environment in which a user can access a server over a network such as the Internet. The medical AI system 20 may also be referred to by other names, such as an action support device.

[0025] The user terminal 30 is a mobile terminal such as a smartphone or wearable device operated by the proposer, and is equipped with a memory, an input interface, a display, a speaker, a communication interface, and a processing circuit (not shown). The user terminal 30 realizes the functions of a smartphone or a wearable device in response to user operations by having the processing circuit execute a program stored in the memory.

[0026] The action suggestion device 40 is a device such as a smartphone or a personal computer operated by a suggestor, and as shown in Fig. 5, includes a memory 41, an input interface 42, a display 43, a speaker 44, a communication interface 45, and a processing circuit 46. The memory 41, the input interface 42, the display 43, the speaker 44, the communication interface 45, and the processing circuit 46 are connected to each other so as to be able to communicate with each other, for example, via a bus.

[0027] Here, the memory 41 is composed of memories for recording electrical information, such as a ROM (Read Only Memory), a RAM (Random Access Memory), a HDD (Hardware Disk Drive), and an image memory, as well as peripheral circuits associated with these memories, such as a memory controller and a memory interface. The memory 41 stores, for example, the program for the action suggestion device 40 and various data, such as various tables, data in the middle of processing, and data after processing. The program is a program for causing a computer to function as the action suggestion device 40 that executes the action suggestion method. The program may be stored in, for example, a non-transitory computer-readable storage medium and distributed, and may be read from the storage medium and installed in the memory 41.

[0028] The input interface 42 may be implemented by a trackball, switch buttons, mouse, keyboard, touchpad (or trackpad) for inputting various instructions, commands, information, selections, and settings from the operator (proposer) into the action proposal device 40, a touch panel display (or touch screen) that integrates a display screen and a touchpad, or the like. The input interface 42 is connected to the processing circuit 46 and converts input operations received from the proposer into electrical signals and outputs them to the processing circuit 46. In this case, the input interface 42 may display a user interface (GUI: Graphical User Interface) on the display 43, allowing the proposer to input various instructions using physical operation components such as a mouse or keyboard. Note that, in this specification, the input interface 42 is not limited to those having physical operation components. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs these electrical signals to the processing circuit 46 is also included as an example of the input interface 42. In the following description, "operation of the input interface 42 by the proposer" is also referred to as "proposer's operation."

[0029] The display 43 is composed of a display main body that displays any data, an internal circuit that supplies display signals to the display main body, and peripheral circuits such as connectors and cables that connect the display main body to the internal circuitry. The display 43 is an example of a display unit and an information output unit.

[0030] The speaker 44 outputs sound in accordance with the output control function 464. The speaker 44 may be a built-in speaker of the action suggestion device 40, or may be an external speaker such as an earphone that can communicate with the action suggestion device 40. The speaker is an example of an audio output unit and an information output unit.

[0031] The communication interface 45 is a circuit for connecting the action suggestion device 40 to a network and performing data communication with other devices. In the following description, the fact that the communication interface 45 is involved in communication between the action suggestion device 40 and other devices will be omitted.

[0032] The processing circuitry 46 reads out programs stored in the memory 41 based on instructions input by the proposer via the input interface 42, and controls the action suggestion device 40 in accordance with the programs. For example, the processing circuitry 46 is a processor that realizes each function of the action suggestion device 40 in accordance with the programs read out from the memory 41. The functions include, for example, an index information acquisition function 461, a relationship information acquisition function 462, a proposal acquisition function 463, and an output control function 464. Note that each function may be realized by distributing it among multiple processors as appropriate. Alternatively, each function or part of each function may be executed by another device as appropriate.

[0033] The index information acquisition function 461 acquires index information that indicates the behavior index of the proposer. The index information acquisition function 461 is an example of an index information acquisition unit.

[0034] The relationship information acquisition function 462 acquires relationship information that indicates the relationship between the proposer and the proposer. The relationship information acquisition function 462 is an example of a relationship information acquisition unit.

[0035] The proposal acquisition function 463 acquires an action proposal for the proposer based on the index information and the relationship information. The proposal acquisition function 463 is an example of a proposal acquisition unit.

[0036] The output control function 464 outputs information based on the suggested action from the display 43 and / or the speaker 44. The display 43, the speaker 44, and the output control function 464 are an example of an information output unit.

[0037] Next, the operation of a system equipped with the action suggestion device configured as above will be explained using the sequence diagrams of Fig. 6 and Fig. 7. First, the operation of registering relationship information will be explained using Fig. 6. Next, the operation of making an action suggestion will be described using Fig. 7. Below, an example will be described in which the user who is the suggestion recipient is a grandfather and the suggester is a grandson.

[0038] First, in step ST11, the user terminal 30 displays a GUI screen for inputting relationship information T1 representing the relationship with the proposer (B), as shown in Figure 3, and transmits the input relationship information T1 to the information storage device 10 through operation by the user (A), who is the proposer.

[0039] After step ST11, in step ST12, the processing circuit 13 of the information storage device 10 stores the received relationship information T1 in the memory 11.

[0040] After step ST12, in step ST13, the action suggestion device 40 displays a GUI screen showing the saved relationship information T1 on the display 43, and adds or modifies the relationship information T1 as appropriate through operation by the proposer (B), and transmits the relationship information T1 to the information storage device 10. Note that the relationship information T1 only needs to be input by at least one of the proposer (A) and the proposer (B).

[0041] After step ST13, in step ST14, the processing circuit 13 of the information storage device 10 stores the received relationship information T1 in the memory 11. This completes the registration of the relationship information T1.

[0042] Next, in step ST21, the user terminal 30 transmits status information indicating the status of the user, who is the proposer (A), to the medical AI system 20. The status information includes, for example, an exercise record indicating the behavioral results. For example, the walking time per day, the number of steps per day, etc. can be used as appropriate as the exercise record.

[0043] After step ST21, in step ST22, when the medical AI system 20 receives the status information, it generates index information for the user (A) based on the status information and transmits the index information to the action suggestion device 40. The index information includes a statement to the effect that "Please exercise moderately every day."

[0044] After step ST22, in step ST23, the processing circuit 46 of the action suggestion device 40 acquires index information of the user (A) and sends a request to the information storage device 10 to send relationship information representing the relationship between the user (A) and the proposer (B).

[0045] After step ST23, in step ST24, when the processing circuit 13 of the information storage device 10 receives a transmission request, it transmits the relationship information "High / Grandchild / Pet (Dog)" representing the relationship between the user (A) and the proposer (B) from the relationship information T1 in the memory 11 to the behavior suggestion device 40.

[0046] After step ST24, in step ST25, the processing circuit 46 of the action suggestion device 40 acquires the relationship information and transmits the user's index information "Get moderate exercise every day" and the relationship information "High / Grandchild / Pet (dog)" to the information storage device 10.

[0047] After step ST25, in step ST26, when the processing circuitry 13 of the information storage device 10 receives the index information and the relationship information, it selects the action proposal "Walk (dog)" from the proposal selection information T2 in the memory 11 based on the index information and the relationship information. Note that in the example of FIG. 8, since the proposal selection information T2 does not include a relationship, the relationship "grandchild" in the relationship information "High / grandchild / pet (dog)" is not used in selecting the action proposal. In addition, the processing circuitry 13 transmits the selected action proposal to the action suggestion device 40.

[0048] After step ST26, in step ST27, when the processing circuitry 46 of the action suggestion device 40 acquires the action suggestion, the processing circuitry 46 outputs the action suggestion as information from the display 43 and the speaker 44. That is, the action suggestion "Walk (dog)" is displayed and output on the display 43 and is also output as audio from the speaker 44, thereby urging the proposer to create a message based on the action suggestion.

[0049] After step ST27, in step ST28, the processing circuit 46, through the operation of the proposer (B), creates a message based on the action proposal, "Grandpa, let's go for a walk with the dog together," and sends the message to the user terminal 30 in chat format.

[0050] After step ST28, in step ST29, the user terminal 30 displays the received message and transmits a response message such as "That's right. Let's go together now" to the action suggestion device 40 in response to an operation by the user (A).

[0051] After that, the user (A) and the proposer (B) walk the dog together, and the user (A) gets moderate exercise.

[0052] As described above, according to one embodiment, the processing circuitry 46 of the action suggestion device 40 acquires index information representing the behavioral index of the suggestion recipient from the medical AI system 20. The processing circuitry 46 acquires relationship information representing the relationship between the suggestion recipient and the proposer from the information storage device 10. The processing circuitry 46 acquires an action suggestion for the suggestion recipient from the information storage device 10 based on the acquired index information and relationship information. The processing circuitry 46 outputs information based on the action suggestion from the display 43 and the speaker 44. In this manner, by acquiring an action suggestion based on index information representing the behavioral index of the suggestion recipient and relationship information representing the relationship between the suggestion recipient and the proposer, an action suggestion can be made that takes into account the relationship between the suggestion recipient and the proposer. For example, an action suggestion that is easily accepted by the user (the suggestion recipient) can be acquired depending on the relationship between the proposer and the user, which is expected to facilitate improvement of the user's behavior. Furthermore, for example, if the user's behavior improves, there is no need to change the proposer, and the same proposer can continue to respond to the user, which is expected to prevent a decrease in user motivation.

[0053] According to an embodiment, the relationship information T1 may include at least one of the intimacy and the relationship between the user (proposer) and the proposer as the relationship. In this case, in addition to the above-described effects, it is possible to make an action proposal taking into consideration at least one of the intimacy and the relationship.

[0054] According to one embodiment, the relationship information T1 may include at least one of the residential relationship, hobby, and habit between the user (recipient) and the proposer as a relationship. In this case, in addition to the above-mentioned effects, it is possible to make an action suggestion taking into consideration at least one of the residential relationship, hobby, and habit.

[0055] (Various variations) The embodiment may be modified as shown in the following modifications, which may be combined with each other.

[0056] (Variation 1) In one embodiment, the suggestion selection information T2 does not include the relationship, but is not limited to this. For example, as shown in Fig. 8, the suggestion selection information T2a may include the relationship between the proposer and the proposer. This not only achieves the effect of the embodiment, but also makes it possible to selectively acquire action suggestions from the suggestion selection information T2a based on the index information and the relationship information, even if the relationship information includes the relationship but does not include the intimacy level and supplemental information.

[0057] (Variation 2) In one embodiment, an action suggestion is made to a user at home with a lifestyle-related disease, but this is not limiting. For example, an action suggestion may be made to a user (suggested person) who is hospitalized due to an illness or injury other than a lifestyle-related disease. In this case, as shown in FIG. 9 , for example, the action indicators in the suggestion selection information T2b represent indicators that the hospitalized patient (suggested person) should observe. Note that the action indicators for lifestyle-related diseases in one embodiment are indicators related to exercise. In contrast, the action indicators for hospitalized patients in Modification 2 include indicators of lifestyle habits other than exercise, such as dietary restrictions and smoking cessation, as well as indicators related to group living, such as not going out without permission and avoiding trouble. Accordingly, for example, in FIG. 9 , the action suggestion in the action suggestion list represents the hospitalized patient's action "changing rooms" in response to the group living behavior indicator, as well as the proposer's actions "introducing XX" and "warning the person" in response to the lifestyle or group living behavior indicator. In this way, Modification 2 achieves the effects of the first embodiment and also allows for the acquisition of action suggestions for the hospitalized user (suggested person).

[0058] In addition, in one embodiment, the higher the intimacy level, the more accompanying behavior suggestions are provided. However, this is not limited to this. For example, in one embodiment, for behavior suggestions related to a behavior indicator for exercise, the higher the intimacy level, the more accompanying behavior is provided. In contrast, in Modification Example 2, for behavior suggestions related to lifestyle habits other than exercise or group living, the higher the intimacy level, the more accompanying behavior is provided. Therefore, even if the intimacy level is high, the behavior suggestions do not necessarily include accompanying behavior. For example, the suggestion selection information T2b shown in FIG. 9 associates a high level of intimacy with “Recommendation for prescription of the . . . app” and “Move room” in the behavior suggestion list. Furthermore, the suggestion selection information T2b associates a medium level of intimacy with “Introduction by . . . . ” and “Instructions by . . . . ” in the behavior suggestion list. Thus, according to Modification Example 2, even if the behavior suggestions in the behavior suggestion list do not include “accompanying,” the higher the intimacy level, the more likely the hospitalized patient (suggested person) is to receive an action suggestion.

[0059] Additionally, the behavioral suggestions associated with a high level of intimacy, such as "recommending a prescription using the app" and "changing rooms," require intrusive communication with the hospitalized patient, but are easily accepted by the hospitalized patient and are expected to facilitate behavioral improvement. Since behavioral suggestions that intrude into the hospitalized patient's inner thoughts are made by a person who is close to the patient, the patient does not feel uncomfortable because the psychological distance between the person who is close to the patient and the behavioral suggestions that reflect a close psychological distance matches. In contrast, if a similar behavioral suggestion is made by a person who is not close to the patient, the patient may feel uncomfortable because the psychological distance between the person who is far from the patient and the behavioral suggestions that reflect a close psychological distance does not match. The feeling of discomfort here includes, for example, the unpleasant impression of receiving a behavioral suggestion from a familiar acquaintance or medical professional who is supposed to be unaware of the user's hobbies, but who is aware of the user's hobbies. However, the feeling of discomfort is not limited to this.

[0060] On the other hand, the behavioral suggestion "Warning..." associated with low intimacy is an act of verbally warning the hospitalized patient or their family, and as with the past, there is a possibility that the hospitalized patient will not accept the behavioral suggestion and their behavior will not improve. However, when a behavioral suggestion in the form of a verbal warning is made by a person with whom the patient is not close, the psychological distance between the suggestor, who is psychologically distant, and the behavioral suggestion, which is psychologically distant, is the same, so it does not make the hospitalized patient or their family feel uncomfortable.

[0061] Furthermore, the suggested actions "Introducing..." and "Instructions...", which are associated with a medium level of intimacy, are intermediate between the two mentioned above, and require some consideration of the hospitalized patient's inner thoughts when conveying them, making them somewhat more acceptable to hospitalized patients.

[0062] In any case, according to the second modification, it is possible to make an action proposal with an appropriate psychological distance depending on the intimacy between the proposer and the proposer, regardless of whether the proposer is accompanied or not. Furthermore, the higher the intimacy, the more likely the proposed action is to be accepted by the hospitalized patient (the proposer).

[0063] (Variation 3) In one embodiment, the medical AI system 20 generates index information based on user status information received from the user terminal 30 and transmits the index information to the action suggestion device 40. However, this is not limiting. For example, in the action suggestion device 40, the index information acquisition function 461 of the processing circuitry 46 may acquire the user's status from the user terminal 30 and acquire index information by analyzing the acquired status. In this case, the index information acquisition function 461 may input status information indicating the user's status into a trained model stored in the memory 41 of the action suggestion device 40 and acquire index information generated by AI analysis from the trained model. Alternatively, the index information acquisition function 461 may store in the memory 41 an analysis table that associates the range of status indicated by the status information with behavioral indices, and acquire index information based on the acquired status information and analysis table. In any case, according to the third modification, in addition to the above-mentioned effects, the external medical AI system 20 can be omitted by configuring the processing circuitry 46 of the action suggestion device 40 to acquire the status of the suggested user and analyze the acquired status to acquire index information.

[0064] (Variation 4) In one embodiment, the registered relationship information is not updated, but this is not limiting. For example, in the action suggestion device 40, the relationship information acquisition function 462 of the processing circuitry 46 may update the relationship information T1 in the memory 11 based on feedback information indicating the proposed user's reaction to the output action suggestion. The feedback information may be information indicating whether the proposed user actually followed the action indicators after the action suggestion. For example, aggregated information obtained by aggregating status information transmitted from the user terminal 30 to the medical AI system 20 over a certain period of time may be used as appropriate. In this case, the medical AI system 20 transmits the status information to the action suggestion device 40 along with index information generated based on the status information. Furthermore, it is preferable for the processing circuitry 46 to update the intimacy level and other information in the relationship information T1 based on the content indicated by the aggregated information over a certain period of time, rather than based on whether a single piece of status information indicates whether the proposed behavior indicators were followed, in order to eliminate biased situations. Furthermore, the action suggestion device 40 receives the status information used to generate the index information from the medical AI system 20, along with the index information. The relationship information acquisition function 462 of the processing circuitry 46 may aggregate the status information received from the medical AI system 20 for a certain period of time, and update the relationship information T1 in the memory 11 based on the obtained aggregated information. Furthermore, the proposal acquisition function 463 of the processing circuitry 46 may update the action proposal based on the index information and the updated relationship information. The output control function 464 of the processing circuitry 46 may output information based on the updated action proposal from the display 43 and / or the speaker 44. According to such a fourth modification, in addition to the effect of the first embodiment, a configuration that updates the relationship information in accordance with the reaction of the proposer to the output action proposal makes it possible to acquire and output an action proposal in accordance with the reaction of the proposer.

[0065] (Variation 5) In the fourth modification of the embodiment, the relationship information T1 is updated using the feedback information, and the action proposal is updated using the updated relationship information T1. However, this is not limiting. For example, in the action proposal device 40, the proposal acquisition function 463 of the processing circuitry 46 may acquire a new action proposal for the suggestion recipient based on feedback information indicating the suggestion recipient's reaction to the output action proposal, index information, and relationship information. The feedback information is as described in the fourth modification. According to the fifth modification, it is possible to acquire an action proposal based on the suggestion recipient's reaction, as in the fourth modification, while omitting the process of updating the relationship information T1 in the fourth modification.

[0066] (Variation 6) In one embodiment, the behavioral suggestion and the behavioral index are different from each other, but are not limited to this. For example, the behavioral suggestion of Modification Example 6 may be information representing a manner of communicating a behavioral index. That is, the behavioral suggestion of Modification Example 6 represents different ways of communicating the same behavioral index, such as "Let's do some moderate exercise together (the proposer is accompanying)," "Let's do some moderate exercise (the proposer is watching)," and "Let's do some moderate exercise (the proposer is not present)." In this example, the same behavioral index means "do some moderate exercise." Different ways of communicating are set in the suggestion selection information, for example, "accompany" is associated with a high level of intimacy, "watch" is associated with a medium level of intimacy, and "not present" is associated with a low level of intimacy. However, the different ways of communicating are not limited to these. According to Modification Example 6, in addition to the effect of the first embodiment, it is possible to obtain a behavioral suggestion in which the way of communicating the behavioral index is adjusted depending on the relationship with the proposer for the same behavioral index.

[0067] (Variation 7) In one embodiment, the index information does not include a relationship, but is not limited to this. For example, the index information of Modification Example 7 may be information that combines a behavioral index and a relationship. That is, the index information of Modification Example 7 matches relationships for the same behavioral index, such as "Let's do moderate exercise with a pet (dog)," "Let's do moderate exercise in the neighborhood," and "Let's do moderate exercise with our shared hobby of golf." In this case, the same behavioral index means "do moderate exercise." The relationships are set in the suggestion selection information, for example, such that the supplemental information "pet (dog)" is associated with a high level of intimacy, the supplemental information "neighborhood" is associated with a medium level of intimacy, and the supplemental information "shared hobby of golf" is associated with a low level of intimacy. According to Modification Example 7, in addition to the effect of the first embodiment, it is possible to obtain, for the same behavioral index, behavioral suggestions that match the relationship depending on the relationship with the suggestee.

[0068] (Variation 8) In one embodiment, the relationship information T1 and the proposal selection information T2 are not updated, but this is not limiting. For example, as shown in FIG. 10 , the processing circuitry 46 of the action suggestion device 40 may include a learning control function 465 in addition to the functions described above. The processing circuitry 46 realizes this learning control function 465 in accordance with a program read from the memory 41, similar to the functions described above. The memory 41 may also store a trained model Md1 controlled by the learning control function 465.

[0069] Here, the learning control function 465 trains the trained model Md1 to generate behavioral indicators, relationships, and behavioral suggestions based on data collected from the network, as shown in Fig. 11. Furthermore, the learning control function 465 updates the relationship information T1 and the suggestion selection information T2 based on the generated behavioral indicators, relationships, and behavioral suggestions.

[0070] The collected data can be used as appropriate, for example, anonymous big data, personal PHR (personal health record) data, and personal history information. Big data is information including general behavioral indicators and behavioral suggestions for lifestyle-related diseases and injuries under various circumstances. Personal PHR data is information related to an individual's health, medical care, etc. Personal history information includes, for example, history information related to communication such as personal emails, SNS (social networking services), and conversations, as well as history information related to events and impressions such as personal diaries (e.g., blogs).

[0071] The trained model Md1 may be additionally trained by the learning control function 465 so as to generate general behavioral indicators, relationships, and action suggestions based on anonymous big data, as shown in Fig. 12. In this case, the learning control function 465 updates the suggestion selection information T2 based on the generated general behavioral indicators, relationships, and action suggestions.

[0072] 13, the learned model Md1 may be additionally learned by the learning control function 465 so as to generate a behavioral index of the proposer, a relationship between the proposer and the proposer, and an action proposal to the proposer, based on the PHR data and history information of an individual. In this case, the learning control function 465 updates the relationship information T1 based on the generated behavioral index, relationship, and action proposal.

[0073] The other configurations are the same as those in the first embodiment.

[0074] According to the above configuration, after the operation shown in FIG. 7 , the processing circuitry 46 of the action suggestion device 40 trains the learned model Md1 to generate behavioral indicators, relationships, and action suggestions based on data collected from the network. The processing circuitry 46 also updates the relationship information T1 and the proposal selection information T2 based on the generated behavioral indicators, relationships, and action suggestions. For example, when anonymous big data is collected, the proposal selection information T2 is updated based on the generated general behavioral indicators, relationships, and action suggestions. Furthermore, for example, when PHR data and personal history information are collected, the relationship information T1 is updated based on the generated behavioral indicators of the suggestee, the relationship between the suggestee and the proposer, and the action suggestions to the suggestee.

[0075] According to such modification 8, even if the behavior indicators, relationships, and content and association of behavior suggestions change over time, the relationship information T1 and the suggestion selection information T2 can be updated.

[0076] (Variation 9) In one embodiment, the action suggestion device 40 is provided as a device separate from the information storage device 10, but this is not limiting. For example, as shown in Fig. 14, an information storage device 10A and a proposer terminal 40A may be provided instead of the information storage device 10 and the action suggestion device 40 in Fig. 1. As shown in Fig. 15, the information storage device 10A includes an action suggestion device 14 therein. The proposer terminal 40A is a mobile terminal that does not include an action suggestion device.

[0077] 15, in the information storage device 10A, the processing circuitry 13 is equipped with, in addition to the management function 131 described above, an index information acquisition function 132, a relationship information acquisition function 133, a proposal acquisition function 134, and an output control function 135. As described above, the index information acquisition function 132, the relationship information acquisition function 133, the proposal acquisition function 134, and the output control function 135 are realized as functions corresponding to a program stored in the memory 11 by the processing circuitry 13 executing the program.

[0078] The index information acquisition function 132 acquires index information that indicates the behavior index of the proposer. The index information acquisition function 132 is an example of an index information acquisition unit.

[0079] The relationship information acquisition function 133 acquires relationship information that indicates the relationship between the proposer and the proposer. The relationship information acquisition function 133 is an example of a relationship information acquisition unit.

[0080] The proposal acquisition function 134 acquires an action proposal for the proposer based on the index information and the relationship information. The proposal acquisition function 134 is an example of a proposal acquisition unit.

[0081] The output control function 135 outputs information based on the action suggestion onto the network via the communication interface 12. The communication interface 12 and the output control function 135 are an example of an information output unit.

[0082] The proposer terminal 40A is a mobile terminal such as a smartphone or tablet terminal operated by the proposer. The proposer terminal 40A is equipped with a memory, an input interface, a display, a speaker, a communication interface, and a processing circuit (not shown). The proposer terminal 40A realizes the functions of a smartphone or tablet terminal in response to user operations by having the processing circuit execute a program in the memory.

[0083] Next, the operation of a system equipped with the action suggestion device configured as above will be described with reference to the sequence diagram of Fig. 16. Note that the operation of registering relationship information is the same as the operation shown in Fig. 6, and therefore the explanation will be omitted. Here, the operation of making action suggestions will be described with reference to Fig. 16.

[0084] Now, as described above, in step ST21, the user terminal 30 transmits status information indicating the status of the user (A) who is the proposer to the medical AI system 20.

[0085] After step ST21, in step ST22A, when the medical AI system 20 receives the status information, it generates index information for the user (A) based on the status information and transmits the index information to the information storage device 10A. The index information includes a statement to the effect that "Please exercise moderately every day."

[0086] After step ST22A, in step ST24, the processing circuit 13 of the information storage device 10A acquires the index information of the user (A), and then reads and acquires the relationship information "High / Grandchild / Pet (Dog)" representing the relationship between the user (A) and the proposer (B) from the memory 11.

[0087] After step ST24A, in step ST26A, the processing circuit 13 of the information storage device 10A selectively acquires the action suggestion "Walk (dog)" from the suggestion selection information T2 in the memory 11 based on the index information "Get moderate exercise daily" and the relationship information "High / Grandchild / Pet (dog)".

[0088] After step ST26A, in step ST27A, the processing circuitry 13 of the information storage device 10A outputs the acquired action proposal onto the network, thereby transmitting the action proposal "walk (dog)" to the proposer terminal 40A.

[0089] After step ST27A, in step ST28-1, when the proposer terminal 40A receives the action proposal, it displays and outputs the action proposal "walk (dog)" on the display and also outputs it as sound from the speaker.

[0090] After step ST28-1, in step ST28-2, the proposer terminal 40A, through the operation of the proposer (B), creates a message based on the action proposal, "Grandpa, let's go for a walk with the dog together," and sends the message to the user terminal 30 in chat format.

[0091] After step ST28-2, in the same manner as described above, in step ST29, the user terminal 30 displays the received message and, in response to the operation of the user (A), transmits a response message such as "That's right. Let's go together now" to the action suggestion device 40.

[0092] After that, the user (A) and the proposer (B) walk the dog together, and the user (A) gets moderate exercise.

[0093] According to variant example 9, in addition to the effects described above, the information storage device 10A is configured to include the action suggestion device 14, so the proposer terminal 40A does not include an action suggestion device, and therefore the proposer terminal 40A can make action suggestions without sharing the load of the action suggestion device.

[0094] According to at least one of the embodiments and modifications described above, an action proposal can be made taking into consideration the relationship between the proposer and the proposer.

[0095] The term "processor" used in the above description refers to a circuit such as a CPU, a GPU, an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)). If the processor is a CPU, for example, the processor realizes its function by reading and executing a program stored in a memory. On the other hand, if the processor is an ASIC, for example, instead of storing a program in a memory, the function is directly incorporated into the circuit of the processor as a logic circuit. Note that each processor in this embodiment is not limited to being configured as a single circuit for each processor, but may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, multiple components in FIG. 1, FIG. 2, FIG. 5, FIG. 14, or FIG. 15 may be integrated into a single processor to realize its function.

[0096] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0097] 10,10A Information storage device 11,41 memory 12,45 Communication Interface 13,46 Processing circuit 131 Management functions 132,461 Index information acquisition function 133,462 Relationship information acquisition function 134,463 Proposal acquisition function 135,464 Output control function 14,40 Action suggestion device 20 Medical AI Systems 30 User terminals 40A Proposer terminal 42 Input Interface 43 Display 44 speakers Md1 pre-trained model T1 Relationship Information T2, T2a, T2b Proposal selection information

Claims

1. an index information acquisition unit that acquires index information representing a behavior index of a proposer; a relationship information acquisition unit that acquires relationship information that indicates a relationship between the proposer and the proposer; a proposal acquisition unit that acquires an action proposal for the proposer based on the index information and the relationship information; an information output unit that outputs information based on the action proposal; An action suggestion device comprising:

2. The action suggestion device according to claim 1 , wherein the index information acquisition unit acquires a state of the suggestee and analyzes the acquired state to acquire the index information.

3. the relationship information acquisition unit updates the relationship information based on feedback information representing a reaction of the proposer to the output action proposal; the suggestion acquisition unit updates the action suggestion based on the index information and the updated relationship information; the information output unit outputs information based on the updated action proposal. The action suggestion device according to claim 1 .

4. The action suggestion device of claim 1 , wherein the suggestion acquisition unit acquires a new action suggestion for the suggestee based on feedback information representing the suggestee's reaction to the output action suggestion, the index information, and the relationship information.

5. The action suggestion device according to claim 1 , wherein the action suggestion is information indicating how to convey the action indicator.

6. The action suggestion device according to claim 1 , wherein the index information is information combining the action index and the relationship.

7. The action suggestion device according to claim 1 , wherein the relationship information includes at least one of a degree of intimacy and a relationship between the proposer and the proposer as the relationship.

8. The action suggestion device according to claim 7 , wherein the relationship information includes at least one of residential relationship, hobby, and habit between the suggestor and the suggestee as the relationship.

9. acquiring, by an index information acquisition unit, index information representing a behavior index of the proposer; acquiring relationship information representing a relationship between the proposer and the proposer by a relationship information acquisition unit; acquiring, by a proposal acquisition unit, an action proposal for the proposer based on the index information and the relationship information; outputting information based on the action proposal by an information output unit; The action suggestion method includes:

10. A function for acquiring indicator information representing the behavioral indicators of the proposer; a function of acquiring relationship information representing the relationship between the proposer and the proposer; a function of acquiring an action proposal for the proposer based on the index information and the relationship information; a function of outputting information based on the action proposal; A program to make the above happen on a computer.

Images