JP7641879B2 - Behavior support device, behavior support method, and behavior support program - Google Patents

Description

The present invention relates to an action support device, an action support method, and an action support program.

There is a demand to extend the healthy lifespan of elderly people, prevent them from transitioning to nursing care, and reduce nursing care costs. Recent cohort studies have revealed that elderly people who are highly involved in society through employment, hobby clubs, sports, etc., or who are highly active in terms of the number of steps taken, walking time, etc. (called social participation behavior), have a longer healthy lifespan and are less likely to develop dementia or transition to nursing care.

Therefore, for example, an information processing device has been proposed that suggests actions that are likely to be performed by a user who tends to be sedentary, based on the user's behavioral history.

JP 2015-49825 A

However, the above-mentioned conventional technology does not take into account the situation the user is in when suggesting actions, so if an action that cannot be carried out is suggested, not only is it ineffective, but it may also lower the user's motivation and cause them to stop using the service.

As there is a wide variation in an individual's lifestyle patterns, health conditions, geographical conditions, etc., it is difficult to take all of these factors into account when proposing actions that an individual can implement. In addition, the situation at the time the action is proposed may not be taken into account, and the timing may not be right, making it impossible to implement the proposed action, such as suggesting to take a walk while at work or to eat after a meal.

The present invention has been made in consideration of the above, and one of its objectives is to suggest effective actions to users.

In order to solve the above-mentioned problems, one aspect of the present invention is a behavior support device that supports a user's behavior to achieve a purpose, and is characterized by having a storage unit that stores behavior pattern data that associates a basic behavior pattern among the user's past behavior patterns with a derived behavior pattern executed by the user following the basic behavior pattern, a behavior measurement unit that measures the user's behavior pattern, and a selection unit that determines whether the user's real-time behavior pattern measured by the behavior measurement unit is similar to the basic behavior pattern among the behavior pattern data stored in the storage unit, selects the derived behavior pattern that corresponds to the basic behavior pattern that is similar to the behavior pattern, and outputs a recommended behavior based on the selected derived behavior pattern via an output device.

According to one aspect of the present invention, for example, it is possible to suggest effective actions to the user.

FIG. 1 is a diagram showing an example of the configuration of an action support apparatus according to a first embodiment. FIG. 4 is a diagram showing an example of the configuration of behavior (location information) data. FIG. 4 is a diagram showing an example of the configuration of user attribute information. FIG. 13 is a diagram showing an example of the configuration of a derived behavior pattern effect evaluation dictionary. FIG. 4 is a diagram showing an example of the configuration of location label information. FIG. 13 is a diagram showing an example of the configuration of a basic behavior pattern list. FIG. 4 is a diagram showing an example of the configuration of behavior pattern data. 6 is a flowchart showing an example of a process for generating behavior pattern data according to the first embodiment. 11 is a flowchart showing an example of a recommendation execution process according to the first embodiment. FIG. 4 is a diagram for explaining a processing example of performing a recommendation according to the first embodiment. FIG. 4 is a diagram for explaining a processing example of performing a recommendation according to the first embodiment. 10 is a flowchart showing an example of a process of updating behavior pattern data according to the first embodiment. 11 is a diagram showing an example of a recommendation notification sent to an action support apparatus by push communication. FIG. 13 is a diagram showing an example of a screen for inquiring about various data. FIG. 13 is a diagram showing an example of a screen displaying the acceptance status of a recommendation. FIG. 13 is a diagram showing an example of a screen displaying the acceptance status of a recommendation. FIG. 13 is a diagram showing an example of a screen displaying the acceptance status of a recommendation. FIG. 11 is a diagram showing an example of the configuration of a derived action pattern effect evaluation dictionary according to the second embodiment. FIG. 11 is a diagram showing an example of the configuration of an action support apparatus and an action support system according to a third embodiment.

Below, an embodiment of the present invention will be described with reference to the drawings. Note that the embodiment described below, including the drawings, is merely an example and does not limit the invention according to the claims. Furthermore, not all of the elements and combinations thereof described in the embodiments are necessarily essential to the solution of the invention. Furthermore, illustrations and descriptions of configurations that are essential to the configuration of the invention but are well known may be omitted.

In the following description, the program may be installed on a device such as a computer, or may be, for example, on a program distribution server or on a computer-readable (e.g., non-transitory) recording medium. Also, in the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

In the following description, the processor may be one or more. The processor is typically a microprocessor such as a CPU (Central Processing Unit), but may be other types of processors such as a GPU (Graphics Processing Unit). The processor may be a single-core or multi-core processor. The processor may be a processor in the broader sense, such as a hardware circuit that performs part or all of the processing (for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit)).

In the following description, identification numbers are used as identification information for various objects, but other types of identification information (e.g., identifiers including letters or symbols) may be used. In the following description, "#Y" represents the number Y, and for example, "XXX#Y" represents XXX identified by the number Y.

In the following description, when elements of the same type are described without distinction, reference signs (or common signs among reference signs) may be used, and when elements of the same type are described with distinction, the identification numbers (or reference signs) of the elements may be used. Also, the number of elements shown in each figure is an example and is not limited to the number shown in the figures.

[Embodiment 1]
1 is a diagram showing an example of the configuration of a behavior support device 1 according to embodiment 1. The behavior support device 1 is a portable general-purpose computer, for example, a slate terminal such as a smartphone or a tablet, or a wearable device such as a smartwatch. The behavior support device 1 includes an auxiliary memory device 10, a main memory device 20, a processor 30, an input device 40, an output device 50, an acceleration sensor 60, and a position information sensor 70.

The primary storage device 20 is a memory that cooperates with the processor 30 when it executes processing. The input device 40 is a device that accepts user input, such as a touch panel. The output device 50 is a device that outputs various information to the user, such as a display or speaker. The acceleration sensor 60 detects the acceleration of the action support device 1. The position information sensor 70 acquires position information of the action support device 1 from a GPS (Global Positioning System), a wireless base station, etc.

The auxiliary memory device 10 is a storage device or the like, and stores behavior (location information) data 11, user attribute information 12, derived behavior pattern effect evaluation dictionary 13, location label information 14, basic behavior pattern list 15, and behavior pattern data 16.

The behavior (location information) data 11 records location information indicating the user's location at each time for each date and day of the week, as shown in the example table configuration in Figure 2.

As shown in the table configuration example in Figure 3, the user attribute information 12 records the age, sex, home location (location information), work days (school days), work location (location information) (school location location information), etc. of the user who owns the action support device 1.

The derived behavior pattern effect evaluation dictionary 13 records the correspondence between the increase or decrease of each corresponding behavior (social participation behavior) in a derived behavior pattern following a basic behavior pattern and the negative and positive evaluations, as shown in the table configuration example in Figure 4. In the example of Figure 4, the derived behavior pattern effect evaluation dictionary 13 provides an interpretation for evaluating the increase or decrease of each behavior in a derived behavior pattern from the perspective of preventing the patient from becoming in a state requiring care. For example, if a derived behavior pattern increases the "number of steps" from a basic behavior pattern, a positive evaluation (+1) is given from the perspective of preventing the patient from becoming in a state requiring care.

In the derived behavior pattern effect evaluation dictionary 13 shown in FIG. 4, the evaluation of changes in behavior (social participation behavior) is done in two stages, negative evaluation and positive evaluation, but it is not limited to this and may be done in three or more stages depending on the degree of increase or decrease. In this case, points are given for each stage, and the derived behavior pattern with the highest total point is selected.

As shown in the table configuration example in Figure 5, the location label information 14 records user-registered labels, POI (Point of Interest) labels, common labels, etc. for each location. User-registered labels provide labels freely edited by the user for the location information of each location. POI labels provide labels of actual names on the map for the location information of each location. Common labels provide labels that generalize user-registered labels or POI labels for the location information of each location. Common labels may be symbols, etc., as long as they are information that can be identified after generalizing the location information of each location.

The basic behavior pattern list 15 records basic behavior patterns, days of the week, places of stay, etc., as shown in the table configuration example in Figure 6. The basic behavior pattern list 15 records the correspondence between days of the week and places of stay, and the basic behavior patterns, which are identification information assigned to these correspondences.

As shown in the table configuration example in FIG. 7, the behavior pattern data 16 records, for each behavior pattern, the frequency (e.g., number of times per month) and behavior content of the basic behavior pattern, as well as the frequency, acceptance, effect, and behavior content of the derived behavior pattern. A behavior pattern is a grouping of records of the behavior (location information) data 11 including basic behavior patterns and derived behavior patterns. Acceptance means that the behavior (content) recommended to the user is accepted and executed by the user. Acceptance is an example of a usefulness score, and may be a cumulative score with +1 for acceptance and -1 for negative acceptance, or an acceptance rate (percentage). Effect is an example of a usefulness score, and indicates the degree to which it is useful for the user to execute the derived behavior pattern in order to achieve a purpose.

The processor 30 is an arithmetic processing device such as a CPU that cooperates with the primary storage device 20 to execute the recommendation generation software 30s. By executing the recommendation generation software 30s, the processor 30 realizes a behavior measurement unit 31, a behavior pattern extraction unit 32, a derived behavior pattern effect evaluation unit 33, a recommendation pattern selection unit 34, and a recommendation pattern execution determination unit 35.

The processor 30 also implements an acceleration sensor control unit 36 and a position information sensor control unit 37 by executing the driver. The acceleration sensor control unit 36 controls the acceleration sensor 60 and acquires acceleration information of the action support device 1. The position information sensor control unit 37 controls the position information sensor 70 and acquires position information of the action support device 1.

The behavior measurement unit 31 measures the behavior of the user who owns the behavior support device 1 based on the acceleration information and location information of the behavior support device 1 acquired in real time, and stores the measurement results in the behavior (location information) data 11 as a behavior history. At this time, the behavior measurement unit 31 determines the behavior (staying at home, movement including walking, staying somewhere other than home, etc.), and assigns location information to stay at a location, and movement or walking to movement between locations. Whether it is movement or walking can be determined from the movement speed based on changes in the location information.

The behavior pattern extraction unit 32 generates a behavior pattern of a character string representing each record (behavior history) of the user's past behavior (location information) data 11, for example, movement with "-" and a stay with the location information of the stay point. If the stay point is the home (point 0), the location information is represented with "." The behavior pattern may look like, for example, "-11111111-." The behavior pattern extraction unit 32 groups (clustering, etc.) the generated behavior patterns.

The behavioral pattern extraction unit 32 also calculates the similarity (behavioral similarity) between all pairs of behavioral patterns in each group. The behavioral similarity can be calculated as a similarity coefficient such as the Dice coefficient or the cost of a shortest path problem such as the Dijkstra algorithm. The behavioral similarity may also be calculated by machine learning.

The behavior pattern extraction unit 32 also extracts, as a basic behavior pattern, the behavior pattern that most frequently has a calculated behavior similarity to other patterns that is equal to or greater than a threshold value. The behavior pattern extraction unit 32 also extracts, as a derived behavior pattern, a behavior pattern that has a behavior similarity to a basic behavior pattern that is equal to or greater than a threshold value.

The behavior pattern extraction unit 32 stores the extracted basic behavior pattern and its day of the week information in the basic behavior pattern list 15. By using the basic behavior pattern list 15 and the location label information 14, it is possible to grasp behavior patterns such as having lunch at a favorite cafe every Saturday (behavior pattern A) and going to work on weekdays (behavior pattern B), etc.

The behavior pattern extraction unit 32 also stores the extracted basic behavior pattern and derived behavior pattern in the behavior pattern data 16. Note that if the same basic behavior pattern and derived behavior pattern already exist in the behavior pattern data 16, they are not stored in the behavior pattern data 16, and the existing data is maintained.

The behavior pattern data 16 has time information of stay at each point according to the storage format of the behavior pattern in the behavior content. For example, a behavior pattern stored in the behavior content in the format of "-11111111-...." is a character string in which the location information of each time is arranged in chronological order, so the position in the character string represents the time information. However, the behavior pattern data 16 does not have day of the week information. Therefore, by referring to the basic behavior pattern list 15, the day of the week information of the basic behavior pattern staying at each point can be obtained. By linking the basic behavior pattern list 15 and the behavior pattern data 16, when detecting a basic behavior pattern from the user's real-time behavior history, a recommendation that can be implemented under a specific situation can be selected taking into account the day of the week and the time. In addition, it is possible to prevent recommendations from being implemented on inappropriate days of the week or at inappropriate times.

The derived behavior pattern effect evaluation unit 33 uses the derived behavior pattern effect evaluation dictionary 13 to evaluate the effect of the derived behavior pattern on the basic behavior pattern extracted by the behavior pattern extraction unit 32. For example, if the derived behavior pattern is a behavior pattern in which the "number of steps" increases from the basic behavior pattern, it is evaluated as a "positive evaluation (+1)", and if it is a behavior pattern in which the number of steps decreases, it is evaluated as a "negative evaluation (-1)". This evaluation is performed for each derived behavior pattern for all behaviors listed in the derived behavior pattern effect evaluation dictionary 13, and the sign of the counted result of "+1" and "-1" (or the counted result itself) is regarded as the effect of the derived behavior pattern.

The derived behavior pattern effect evaluation unit 33 stores the evaluation results of the effect of the derived behavior pattern in the behavior pattern data 16.

The recommendation pattern selection unit 34, in the same manner as the behavioral pattern extraction unit 32, generates behavioral patterns of character strings that represent, for example, movement with "-", home (point 0) with "・", and stay with the location information of the stay point, from the real-time behavioral history measured and stored in the behavioral (location information) data 11.

The recommendation pattern selection unit 34 also calculates the behavioral similarity between the behavioral pattern of the real-time behavior history and the basic behavioral pattern stored in the behavioral pattern data 16. When calculating the behavioral similarity, time conditions such as the time difference between the behavioral patterns being within a predetermined time are also taken into consideration.

The recommendation pattern selection unit 34 selects a derived behavior pattern with a positive acceptance rate, the top β% (β is a predetermined number) and an effect of "+" from among derived behavior patterns of basic behavior patterns with behavior similarity in the top α% (α is a predetermined number). At this time, the recommendation pattern selection unit 34 selects a derived behavior pattern whose day of the week (basic behavior pattern list 15 (Figure 6)) of the behavior pattern corresponding to the basic behavior pattern matches the day of the week in the real-time history. If the day of the week of the behavior pattern corresponding to the basic behavior pattern does not match the day of the week in the real-time history, for example, there may be an inconvenience in which it is detected that the person is working on a Sunday, which is a holiday. Then, the recommendation pattern selection unit 34 recommends the selected derived behavior pattern as a recommended pattern to the user via the output device 50.

The recommendation pattern execution determination unit 35 performs a similarity search of behavioral patterns based on the actual behavioral history after the recommendation was made by the recommendation pattern selection unit 34, and searches for behavioral patterns similar to the recommendation pattern implemented by the recommendation pattern selection unit 34. From the actual behavioral history stored in the behavior (location information) data 11, the recommendation pattern execution determination unit 35 generates, in the same manner as the behavioral pattern extraction unit 32, character string patterns that represent, for example, movement after the recommendation with "-", home (point 0) with "・", and stay with location information of the stay point. When a behavioral pattern whose behavioral similarity with the recommendation pattern selected by the recommendation pattern selection unit 34 is γ (γ is a predetermined number) or more is hit by the similarity search, the recommendation pattern execution determination unit 35 adds "+1" to the acceptance of the corresponding recommendation pattern.

(Behavior Pattern Data Generation Process According to the First Embodiment)
8 is a flowchart showing an example of a process of generating behavior pattern data according to embodiment 1. The process of generating behavior pattern data is executed at a predetermined cycle or at the timing of a user instruction.

First, in step S11, the behavior pattern extraction unit 32 reads each record (behavior history), location information, and acceleration information of the behavior (location information) data 11. Next, in step S12, the behavior pattern extraction unit 32 generates and groups character string patterns representing each behavior history read in step S11.

Next, in step S13, the behavior pattern extraction unit 32 calculates the similarity (behavior similarity) for all pairs of patterns in each group grouped in step S12. Next, in step S14, the behavior pattern extraction unit 32 extracts the pattern that most frequently has a behavior similarity to other patterns calculated in step S13 that is equal to or greater than a threshold value as a basic behavior pattern of the group. Next, in step S15, the behavior pattern extraction unit 32 extracts the pattern that has a behavior similarity to the basic behavior pattern that is equal to or greater than a threshold value as a derived behavior pattern.

Next, in step S16, the derived behavior pattern effect evaluation unit 33 uses the derived behavior pattern effect evaluation dictionary 13 to evaluate the effect of the derived behavior pattern on the basic behavior pattern extracted by the behavior pattern extraction unit 32. Next, in step S17, the derived behavior pattern effect evaluation unit 33 stores the evaluation results of the extracted basic behavior pattern, derived behavior pattern, and the effect of the derived behavior pattern in the behavior pattern data 16.

(Recommendation Execution Process According to the First Embodiment)
9 is a flowchart illustrating an example of a recommendation implementation process according to embodiment 1. The recommendation implementation process is executed in real time.

First, in step S21, the recommendation pattern selection unit 34 reads the behavior history, location information, and acceleration information that are measured in real time and stored in the behavior (location information) data 11. Next, in step S22, the recommendation pattern selection unit 34 generates a behavior pattern of a character string that represents the behavior history read in step S21.

Next, in step S23, the recommendation pattern selection unit 34 performs a similarity search of the basic behavior patterns stored in the behavior pattern data 16, and searches for basic behavior patterns that are similar to the behavior patterns in the real-time behavior history.

Next, in step S24, the recommendation pattern selection unit 34 selects a derived behavior pattern with a positive acceptance rate, in the top β% (β is a predetermined number), and a "+" effect from among derived behavior patterns of basic behavior patterns with a time difference within a predetermined time and behavior similarity in the top α%. At this time, the recommendation pattern selection unit 34 selects a derived behavior pattern whose day of the week (basic behavior pattern list 15 (Figure 6)) of the behavior pattern corresponding to the basic behavior pattern matches the day of the week in the real-time history.

Next, in step S25, the recommendation pattern selection unit 34 sets the derived behavior pattern selected in step S24 as a recommendation pattern, and recommends the recommendation pattern to the user via the output device 50.

Figures 10 and 11 are diagrams for explaining a processing example of recommendation implementation according to embodiment 1.

Figure 10 (a) shows a behavior pattern (behavior pattern A) from the behavior history of "Leaved home (point 0) at 11:00, walked for about 30 minutes to point 1, stayed at point 1 for 90 minutes, and returned home after walking for about 30 minutes." While behavior pattern A appears frequently, assume that behavior pattern A' exists in the past as a derived behavior pattern of behavior pattern A, which stops at point 2, as shown in Figure 10 (b). When the situation during the execution of behavior pattern A is predicted or detected, if a visit to point 2 is recommended, as shown in Figure 10 (c), an increase in the number of points stayed (visited) and walking time can be expected. By making a recommendation that has been realized in the past under a situation similar to that at the time of the past, a recommendation with a high feasibility can be implemented.

Figure 11 (a) shows a behavior pattern (behavior pattern B) in the behavior history of "commuting (round trip) by approximately 15 minutes (e.g., train)". While behavior pattern B appears frequently, assume that behavior pattern B', which was derived from behavior pattern B in the past and involved walking home without taking the train on the return journey, exists as shown in Figure 11 (b). When a situation in which behavior pattern B is being performed is predicted or detected, as shown in Figure 11 (c), if behavior pattern B', which involves walking home on the return journey, is recommended, an increase in walking time can be expected. The recommendation of behavior pattern B' is performed only when external information such as the probability of precipitation and temperature obtained from a weather forecast corresponds to a situation in which it is possible to perform behavior pattern B', and therefore the recommendation can be performed under a situation in which it is feasible.

(Action Pattern Data Update Process According to the First Embodiment)
12 is a flowchart showing an example of a process of updating behavior pattern data according to embodiment 1. The process of updating behavior pattern data is executed at a predetermined interval or at the timing of a user instruction.

First, in step S31, the recommendation pattern execution determination unit 35 performs a similarity search of behavioral patterns based on actual behavioral history after the recommendation was made by the recommendation pattern selection unit 34, and searches for patterns similar to the recommendation pattern implemented by the recommendation pattern selection unit 34.

Next, in step S32, the recommendation pattern execution determination unit 35 determines whether or not a behavior pattern having a behavior similarity of γ or more with the recommendation pattern implemented by the recommendation pattern selection unit 34 exists among the patterns based on the actual behavior history after the recommendation is implemented. If a behavior pattern having a behavior similarity of γ or more with the recommendation pattern implemented by the recommendation pattern selection unit 34 exists among the patterns based on the actual behavior history after the recommendation is implemented (step S32 YES), the recommendation pattern execution determination unit 35 proceeds to step S33, and if no behavior pattern exists (step S32 NO), the behavior pattern data update process ends.

In step S33, the recommendation pattern execution determination unit 35 updates the behavioral pattern data 16 by, for example, incrementing the acceptance rate of the corresponding recommendation pattern implemented by the recommendation pattern selection unit 34 by “+1.”

(Example of screen display of action support device 1)
Next, various screen displays displayed on the screen of the output device 50 of the action support apparatus 1 will be described with reference to FIGS.

Fig. 13 is a diagram showing an example of a recommendation notification by push communication notified to the action support device 1. In step S25 of the recommendation implementation process (Fig. 9), the contents of the recommendation are output in chronological order by push notification on the display screen 50A of the output device 50 of the action support device 1, as shown in Fig. 13. For example, push notification 51A1 has contents that recommend the optimal action to be taken subsequently when a certain action pattern of the user is predicted, such as "Action pattern A of staying at point 1 has been detected. Why don't you extend your trip to point 2?", and corresponds to Fig. 10(c). Also, for example, push notification 51A2 has contents that say "Action pattern B of staying at point n has been detected. Why don't you walk home today?", and corresponds to Fig. 11(c).

Figure 14 is a diagram showing an example of a screen for querying various data. When the behavioral data display button 52A1 is tapped, behavioral (location information) data 11 (Figure 2) is displayed in the display pane 52A of the output device 50 of the behavior support device 1. When the registered behavioral pattern display button 52A2 is tapped, a basic behavioral pattern list 15 (Figure 6) is displayed in the display pane 52A. When the recommendation acceptance status display button 52A3 is tapped, a recommendation acceptance status list (Figure 15) is displayed in the display pane 52A. When the label edit button 52A4 is tapped, location label information 14 (Figure 5) is displayed in an editable format in the display pane 52A.

In the example shown in FIG. 15, the recommendation acceptance status includes the date of recommendation, the recommendation content, and the result (whether the recommendation was accepted or not) displayed for each recommendation. The recommendation acceptance status list is not limited to the example shown in FIG. 15, and may display the acceptance rate for each day in a calendar format as in FIG. 16. Alternatively, as in FIG. 17, the number of recommendations, the number of acceptances, and the number of acceptances/number of recommendations=acceptance rate may be displayed for each recommendation content.

According to the above-described first embodiment, by referring to the user's past actions, it is possible to recommend highly feasible behavioral patterns that have been performed by the user in similar circumstances in the past. In addition, by learning the user's receptivity to recommendations, it is possible to take into account feasibility factors that are not reflected in the user's situation or actions (e.g., wanting to go straight home after work because one is tired, or being highly motivated to take a walk after eating out). As a result, by effectively visualizing the social participation actions of users such as elderly people, it is possible to encourage behavioral change in users through intervention services via the action support device 1, thereby promoting social participation actions, which in turn can prevent elderly people from becoming dependent on care and maintain the soundness of the long-term care insurance system.

[Embodiment 2]
In the first embodiment, the action support device 1 recommends actions that will prevent the user from becoming dependent on care. In contrast, in the second embodiment, the action support device 1 recommends actions that will lead to the user's successful diet. In the second embodiment, the action support device 1 stores a derived action pattern effect evaluation dictionary 13C in the auxiliary storage device 10 instead of the derived action pattern effect evaluation dictionary 13.

Figure 18 is a diagram showing an example of the configuration of the derived behavior pattern effect evaluation dictionary 13C according to the second embodiment. The derived behavior pattern effect evaluation dictionary 13C differs from the derived behavior pattern effect evaluation dictionary 13 of the first embodiment in that it provides an interpretation for evaluating the increase or decrease of each behavior of a derived behavior pattern from the viewpoint of succeeding in a diet. With reference to Figure 18, for example, if a derived behavior pattern causes "dinner time" to be after 8 p.m. compared to the basic behavior pattern, it is understood that a negative evaluation (-1) is given from the viewpoint of succeeding in a diet.

If the purpose of the action support device 1 is different, the evaluation may change even for the same action. By adopting an appropriate derived action pattern effect evaluation dictionary that is in line with the purpose of the action support device 1, the disclosed technology can be applied to all fields, not just the healthcare field such as nursing care and dieting in embodiment 1.

[Embodiment 3]
In the first embodiment, the action support device 1 recommends an action of the user by using the action pattern data 16 closed in each action support device 1. In contrast, in the third embodiment, the action support device 1 recommends an action by also using the action pattern data 16 of other action support devices 1. Fig. 19 is a diagram showing an example of the configuration of the action support device 1 and the action support system S according to the third embodiment. In Fig. 19, illustration is omitted as appropriate.

The action support system S includes a server 100 and an action support device 1 for each of a plurality of users connected via a network. In order to share the action pattern data 16, the action support devices 1 transmit the position label information 14 of each action support device 1 together with the action pattern data 16 of each action support device 1 to the server 100. Alternatively, the action support devices 1 convert the position information included in the action pattern of the action pattern data 16 of each action support device 1 into a common label using the position label information 14 before transmitting it.

The server 100 accumulates the location label information 14 and behavior pattern data 16 received from each behavior support device 1. Then, each behavior support device 1 detects a behavior pattern when the behavior pattern of the real-time behavior history detected by the device itself is similar to a basic behavior pattern stored in the behavior pattern data 16 on the device itself or on the server 100. Then, an appropriate derived behavior pattern is selected from the derived behavior patterns corresponding to the basic behavior pattern stored in the behavior pattern data 16 on the device itself or on the server 100, and the appropriate derived behavior pattern is recommended to the user as a recommended pattern.

Each behavior support device 1 can interpret the location information of other behavior support devices 1 using the common label in the location label information of the other behavior support devices 1 as a clue, and can use the behavior pattern data 16 of the other behavior support devices 1 as information that is also meaningful to its own device.

In addition, multiple action support devices 1 may directly exchange and share location label information 14 and action pattern data 16 between each other without going through the server 100.

The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and are not necessarily limited to those having all of the configurations described. Furthermore, as long as there is no contradiction, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, replace, integrate, or distribute part of the configuration of each embodiment. Furthermore, the configurations and processes shown in the embodiments can be appropriately distributed, integrated, or replaced based on processing efficiency or implementation efficiency.

1: Behavior support device, 10: Auxiliary storage device, 11: Behavior (location information) data, 12: User attribute information, 13, 13C: Derived behavior pattern effect evaluation dictionary, 14: Location label information, 15: Basic behavior pattern list, 16: Behavior pattern data, 20: Main storage device, 30: Processor, 31: Behavior measurement unit, 32: Behavior pattern extraction unit, 33: Derived behavior pattern effect evaluation unit, 34: Recommended pattern selection unit, 35: Recommended pattern execution determination unit, 100: Server

Claims (12)

An action support device that supports an action of a user,
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
a behavior measurement unit that measures a behavior pattern of the user;
a selection unit that determines whether or not the real-time behavior pattern of the user measured by the behavior measurement unit is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit, selects the derived behavior pattern corresponding to the basic behavior pattern similar to the real-time behavior pattern, and outputs a recommendation of a behavior based on the selected derived behavior pattern via an output device ,
the storage unit stores an acceptability score indicating a degree to which a recommended behavior based on the derived behavior pattern is output to the user and the recommended behavior is accepted and executed by the user in association with the derived behavior pattern;
The behavior support device, wherein the selection unit selects the derived behavior pattern based on the acceptability score. The action support device according to claim 1 ,
an execution determination unit that refers to a history of the user's behavioral patterns measured by the behavior measurement unit, determines whether the recommendation for the behavior has been accepted by the user and the behavior has been performed, and updates the acceptability score in accordance with the determination result. An action support device that supports an action of a user,
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
a behavior measurement unit that measures a behavior pattern of the user;
a selection unit that determines whether or not the real-time behavior pattern of the user measured by the behavior measurement unit is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit, selects the derived behavior pattern corresponding to the basic behavior pattern similar to the real-time behavior pattern, and outputs a recommendation of a behavior based on the selected derived behavior pattern via an output device ,
the storage unit stores a usefulness score indicating a degree of usefulness of the derived behavior pattern with respect to achievement of a purpose when the derived behavior pattern is executed by the user in association with the derived behavior pattern;
The action support device, wherein the selection unit selects the derived action pattern based on the usefulness score. The action support device according to claim 3 ,
the storage unit further stores an evaluation dictionary for evaluating the derived behavior pattern executed subsequent to the basic behavior pattern in accordance with the purpose;
The action support device, wherein the usefulness score is determined based on the evaluation dictionary. The action support device according to claim 4 ,
The purpose is to prevent the user from transitioning to a care-requiring state,
The action support device, wherein the evaluation dictionary is a dictionary for evaluating whether or not the derived action pattern is useful for preventing the user from transitioning to a care-requiring state. An action support device that supports an action of a user,
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
a behavior measurement unit that measures a behavior pattern of the user;
a selection unit that determines whether or not the real-time behavior pattern of the user measured by the behavior measurement unit is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit, selects the derived behavior pattern corresponding to the basic behavior pattern similar to the real-time behavior pattern, and outputs a recommendation of a behavior based on the selected derived behavior pattern via an output device ,
an extraction unit that groups the user's past behavior patterns measured by the behavior measurement unit, extracts, within each group, a behavior pattern whose similarity to the most other behavior patterns satisfies a predetermined condition as the basic behavior pattern, extracts the other behavior patterns whose similarity to the basic behavior pattern satisfies a predetermined condition as the derived behavior pattern, and stores in the storage unit the behavior pattern data in which the extracted basic behavior patterns and the derived behavior patterns are associated with each other. The action support device according to any one of claims 1 to 6 ,
The selection unit is
11. A behavior support device, comprising: a behavior support apparatus, which searches for the behavior pattern data stored in the storage unit of the apparatus itself as well as the behavior pattern data of other behavior support apparatuses and selects the derived behavior pattern. The action support device according to any one of claims 1 to 6 ,
The selection unit is
and selecting the derived behavior pattern executable by the user based on a situation including time information or location information indicated by a real-time behavior pattern of the user measured by the behavior measurement unit. An action support method executed by an action support device that supports a user 's action, comprising:
The action support device includes:
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
The action support device,
Measuring the user's behavioral patterns;
determining whether or not the measured real-time behavior pattern of the user is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit;
Selecting the derived behavior pattern corresponding to the basic behavior pattern similar to the behavior pattern;
and outputting to the user a recommended behavior based on the selected derived behavior pattern ,
the storage unit stores an acceptability score indicating a degree to which a recommended behavior based on the derived behavior pattern is output to the user and the recommended behavior is accepted and executed by the user in association with the derived behavior pattern;
The behavior support device selects the derived behavior pattern based on the acceptability score.
The action support method further comprises the steps of: An action support method executed by an action support device that supports a user's action, comprising:
The action support device includes:
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
The action support device,
Measuring the user's behavioral patterns;
determining whether or not the measured real-time behavior pattern of the user is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit;
Selecting the derived behavior pattern corresponding to the basic behavior pattern similar to the behavior pattern;
A behavior recommendation based on the selected derived behavior pattern is output to the user.
Each process has
the storage unit stores a usefulness score indicating a degree of usefulness of the derived behavior pattern with respect to achievement of a purpose when the derived behavior pattern is executed by the user in association with the derived behavior pattern;
The behavior support device selects the derived behavior pattern based on the usefulness score.
The action support method further comprises the steps of : An action support method executed by an action support device that supports a user's action, comprising:
The action support device includes:
a storage unit that stores behavior pattern data in which a basic behavior pattern among the user's past behavior patterns is associated with a derived behavior pattern executed by the user subsequent to the basic behavior pattern;
The action support device,
Measuring the user's behavioral patterns;
determining whether or not the measured real-time behavior pattern of the user is similar to the basic behavior pattern of the behavior pattern data stored in the storage unit;
Selecting the derived behavior pattern corresponding to the basic behavior pattern similar to the behavior pattern;
A behavior recommendation based on the selected derived behavior pattern is output to the user.
Each process has
The action support device,
grouping the measured past behavioral patterns of the user, and extracting, from each group, a behavioral pattern that satisfies a predetermined condition in terms of similarity with the most other behavioral patterns as the basic behavioral pattern;
extracting the other behavior patterns whose similarity to the basic behavior pattern satisfies a predetermined condition as the derived behavior pattern;
The behavior pattern data in which the extracted basic behavior pattern and the derived behavior pattern are associated with each other is stored in the storage unit.
The action support method further comprises the steps of: A behavior support program for causing a computer to function as the behavior support device according to any one of claims 1 to 8 .

Images