Life rhythm estimation device and life rhythm estimation program

Abstract

To provide a technique capable of generalizing and grasping the life rhythm of an object person.SOLUTION: The system includes a data acquisition unit (11) that acquires daily behavior-related data related to a daily behavior of a subject detected by a sensor (2), a daily behavior specifying unit (12) that compares the daily behavior-related data acquired by the data acquisition unit (11) with a plurality of pieces of basic pattern data respectively indicating basic patterns of mutually different life patterns, and specifies a life pattern of the subject in a life pattern specifying period, a life pattern aggregation unit (13) that aggregates the life pattern of the subject in a life rhythm estimation period based on the life pattern of the subject specified by the life pattern specifying unit (12) and a life pattern aggregation condition, and a life rhythm estimation unit (14) that estimates a life rhythm of the subject from an aggregation result of the life pattern of the subject aggregated by the life pattern aggregation unit (13).SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a life rhythm estimation device and a life rhythm estimation program. [Background technology]

[0002] It is known that problems with one's lifestyle or habits (hereinafter collectively referred to as "lifestyle") can cause problems with one's physical and mental health. BACKGROUND ART Conventionally, there is a technique for observing the behavior of a person to be observed (hereinafter referred to as a "subject") in their daily life and providing the subject with data on lifestyle improvements, etc. For example, as a technology for detecting abnormal behavior of a subject, Patent Document 1 discloses an information processing device that generates a behavior model of the subject based on a behavior history in which past location information is associated with time, and determines abnormal behavior of the subject based on the behavior model and location information. In this information processing device, a plurality of behavior patterns that associate the subject's movement route with the time period of movement are generated as a behavior model, and the information processing device determines that the subject's behavior based on the location information is abnormal if it differs from a selected behavior pattern selected from the plurality of behavior patterns. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-73858 Summary of the Invention [Problem to be solved by the invention]

[0004] Although a subject's lifestyle rhythm is unique to that subject, it would be useful to have a technology that can generalize and understand the lifestyle rhythm of such a subject. By generalizing and understanding the subject's lifestyle rhythm, it would be possible to determine whether the lifestyle rhythm is good or bad from a general perspective that is common to an unspecified number of people. However, there has been a problem in that such a technology has not been provided. For example, even in the conventional technology disclosed in Patent Document 1, the behavioral pattern of a subject is generated based only on the subject's own behavioral history and is not generalized. Therefore, such conventional technology cannot determine whether the subject's behavior is good or bad from a general perspective, and the above-mentioned problem cannot be solved.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a technology that can generalize and grasp the life rhythm of a subject. [Means for solving the problem]

[0006] The lifestyle rhythm estimation device disclosed herein includes a data acquisition unit that acquires daily activity-related data regarding the daily activities of a subject detected by a sensor; a lifestyle pattern identification unit that compares the daily activity-related data acquired by the data acquisition unit with a plurality of basic pattern data that each indicate different basic patterns of lifestyle patterns to identify the subject's lifestyle pattern during a period used for lifestyle pattern identification; a lifestyle pattern aggregation unit that aggregates the subject's lifestyle pattern during the lifestyle rhythm estimation period based on the subject's lifestyle pattern identified by the lifestyle pattern identification unit and lifestyle pattern aggregation conditions that set an aggregation unit and aggregation method for the subject's lifestyle pattern during the lifestyle rhythm estimation period; and a lifestyle rhythm estimation unit that estimates the subject's lifestyle rhythm from the aggregation results of the subject's lifestyle pattern aggregated by the lifestyle pattern aggregation unit. [Effects of the Invention]

[0007] According to the present disclosure, the lifestyle rhythm estimation device is configured as described above, and therefore can generalize and grasp the lifestyle rhythm of a subject. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram illustrating an example of the configuration of a lifestyle rhythm estimation device according to a first embodiment. [Figure 2] 2A shows an example of a basic pattern in embodiment 1, where FIG. 2A shows a basic pattern in which the refrigerator door is opened and closed in the morning and evening, FIG. 2B shows a basic pattern in which the refrigerator door is opened and closed in the morning, afternoon, and evening, FIG. 2C shows a basic pattern in which the refrigerator door is opened and closed only in the afternoon, and FIG. 2D shows a basic pattern in which the refrigerator door is not opened or closed in a day. [Figure 3] FIG. 2 is a diagram for explaining an example of a result of tallying lifestyle patterns tallyed by a lifestyle pattern tallying unit in the first embodiment. [Figure 4] FIG. 2 is a diagram for explaining an example of detection of a change in a life rhythm by a change detection unit in the first embodiment. [Figure 5] 4 is a flowchart for explaining the operation of a lifestyle pattern identification process in the lifestyle rhythm estimation device according to the first embodiment. [Figure 6] 4 is a flowchart for explaining the operation of a life rhythm estimation process in the life rhythm estimation device according to the first embodiment. [Figure 7] FIG. 10 is a diagram for explaining another example of the result of tallying the lifestyle patterns tallyed by the lifestyle pattern tallying unit in the first embodiment. [Figure 8] FIG. 10 is a diagram illustrating another example of detection of a change in a life rhythm by the change detection unit in the first embodiment. [Figure 9] FIG. 10 is a diagram for explaining another example of the result of tallying the lifestyle patterns tallyed by the lifestyle pattern tallying unit in the first embodiment. [Figure 10]FIG. 10 is a diagram illustrating another example of detection of a change in a life rhythm by the change detection unit in the first embodiment. [Figure 11] 11A is a diagram illustrating an example of a method for detecting a change in a subject's lifestyle rhythm based on the ratio of each lifestyle pattern in the aggregation results of multiple lifestyle patterns, which are indicated by the subject's lifestyle rhythm estimated by the change detection unit over multiple time-series lifestyle rhythm estimation periods in embodiment 1. FIG. 11A is a diagram illustrating an example of a method for the change detection unit to detect a change in a subject's lifestyle rhythm from a transition in the ratio of each lifestyle pattern. FIG. 11B is a diagram illustrating an example of a method for the change detection unit to detect a change in a subject's lifestyle rhythm from a sudden change in the ratio of each lifestyle pattern. FIG. 11C is a diagram illustrating an example of a method for the change detection unit to detect a change in a subject's lifestyle rhythm taking into account the ratio and variation of each lifestyle pattern. [Figure 12] 12A and 12B are diagrams illustrating an example of a hardware configuration of a lifestyle rhythm estimation device according to Embodiment 1. As shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0010] Embodiment 1 The lifestyle rhythm estimation device according to the first embodiment estimates the lifestyle rhythm of a person to be observed (hereinafter referred to as "subject") so that the lifestyle rhythm can be generalized and understood. In the first embodiment, the life rhythm is estimated by determining how often each life pattern is repeated. Here, a life pattern is a generalized type that indicates what kind of behavior a subject exhibits during a certain period of time (hereinafter referred to as a "first period") in daily life. Hereinafter, behavior in daily life is referred to as daily behavior. Daily behavior includes activities such as eating, sleeping, working, or exercising, as well as rest. In embodiment 1, more specifically, the life rhythm is estimated by aggregating each life pattern within a certain period of time (hereinafter referred to as a "second period") that includes multiple of the above-mentioned first periods. The first period is a relatively short period such as "one day," and the second period is a relatively long period such as "one month."

[0011] Here, a person does not always live according to a fixed lifestyle pattern during each of the plurality of first periods, but lives according to various lifestyle patterns. In other words, there can be various patterns even within a generalized lifestyle pattern. Taking these factors into consideration, the lifestyle rhythm estimation device according to the first embodiment identifies the lifestyle pattern of a subject during a first period based on which of a plurality of generalized lifestyle patterns (hereinafter referred to as "basic patterns") the subject's lifestyle pattern during the first period corresponds to, and estimates the subject's lifestyle rhythm during the second period from the identified lifestyle pattern of the subject during the first period. This enables the lifestyle rhythm estimation device to estimate a generalized lifestyle rhythm.

[0012] In the following first embodiment, the first period is referred to as a "period for identifying a lifestyle pattern," and the second period is referred to as a "period for estimating a lifestyle rhythm." In the following first embodiment, as an example, the period for identifying a lifestyle pattern is set to one day, and the period for estimating a lifestyle rhythm is set to one month.

[0013] In the following first embodiment, as an example, the subject's daily behavior, which is the subject's target for estimating a lifestyle rhythm, is assumed to be eating. That is, the lifestyle rhythm estimation device estimates the subject's lifestyle rhythm based on the meals in the subject's daily life. In the following first embodiment, for example, it is assumed that when the refrigerator door is opened or closed, it can be assumed that the subject has eaten a meal.

[0014] FIG. 1 is a diagram showing an example of the configuration of a lifestyle rhythm estimation device 1 according to the first embodiment. The life rhythm estimation device 1 is connected to a sensor 2 and an output device 3 .

[0015] The sensor 2 detects the subject's daily behavior and outputs data related to the detected daily behavior of the subject (hereinafter referred to as "daily behavior-related data") to the lifestyle rhythm estimation device 1. The daily behavior-related data includes data indicating that the daily behavior occurred and data indicating the date and time when the daily behavior was detected. The daily behavior-related data may also include data indicating what kind of daily behavior was performed. Here, sensor 2 is provided in, for example, a refrigerator and detects whether the refrigerator door is open or closed. When sensor 2 detects whether the refrigerator door is open or closed, it outputs daily activity-related data to lifestyle rhythm estimation device 1, including data indicating that the refrigerator door was opened or closed and data indicating the date and time when the refrigerator door was detected to be opened or closed. Sensor 2 may be any sensor that can detect whether the refrigerator door is open or closed, but here, as an example, sensor 2 is a well-known sensor that uses magnetism to detect whether the door is open or closed based on whether a circuit is closed or not. Sensor 2 takes on two values, for example, 0 or 1. The sensor 2 outputs data indicating that the refrigerator door has been opened or closed to the lifestyle rhythm estimation device 1 as daily activity related data.

[0016] The life rhythm estimation device 1 acquires daily activity related data from the sensor 2, and estimates the life rhythm of the subject based on the acquired daily activity related data. When the lifestyle rhythm estimation device 1 estimates the lifestyle rhythm of the subject, it generates data relating to the estimated lifestyle rhythm of the subject (hereinafter referred to as "provided data") and outputs it to the output device 3. The lifestyle rhythm estimation device 1 may store the provided data in its own storage unit 17 or in an external storage device (not shown). A detailed configuration example of the lifestyle rhythm estimation device 1 will be described later.

[0017] The output device 3 outputs the provided data output from the life rhythm estimation device 1. The output device 3 is, for example, a display device that displays the provided data or an audio output device that outputs the provided data as audio. The output device 3 is provided in, for example, a mobile terminal such as a smartphone or tablet terminal held by the administrator or the subject, or a PC (Personal Computer). An example of the provided data output from the life rhythm estimation device 1 will be described later in detail.

[0018] The life rhythm estimation device 1 is provided in, for example, a server (not shown). As shown in FIG. 1, the lifestyle rhythm estimation device 1 includes a data acquisition unit 11, a lifestyle pattern identification unit 12, a lifestyle pattern aggregation unit 13, a lifestyle rhythm estimation unit 14, a change detection unit 15, and a data provision unit 16. In embodiment 1, the lifestyle rhythm estimation device 1 performs a "lifestyle pattern identification process" that identifies the lifestyle pattern of a subject based on daily activity-related data output from the sensor 2, and a "lifestyle rhythm estimation process" that estimates the lifestyle rhythm of the subject based on the lifestyle pattern of the subject identified by the "lifestyle pattern identification process." In the lifestyle rhythm estimation device 1, the data acquisition unit 11 and lifestyle pattern identification unit 12 function in a "lifestyle pattern identification process." In the lifestyle rhythm estimation device 1, the lifestyle pattern aggregation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, and data provision unit 16 function in a "lifestyle rhythm estimation process."

[0019] The data acquisition unit 11 acquires daily activity related data from the sensor 2 . The data acquisition unit 11 outputs the acquired daily activity related data to the lifestyle pattern identification unit 12. Furthermore, the data acquiring unit 11 stores the acquired daily activity related data in the storage unit 17.

[0020] The lifestyle pattern identification unit 12 compares the daily behavior related data acquired by the data acquisition unit 11 with multiple data (hereinafter referred to as "basic pattern data") each showing a different basic pattern, and identifies the subject's lifestyle pattern identification period, in this case, one day's lifestyle pattern. In this case, the basic patterns are a plurality of generalized types of opening and closing of the refrigerator door during one day, which is the period for identifying a lifestyle pattern.

[0021] Here, the basic pattern will be described using a specific example. FIG. 2 is a diagram illustrating an example of a basic pattern according to the first embodiment. Figure 2 shows four basic patterns, which are generalized types of refrigerator door opening and closing over the course of a day.

[0022] The basic patterns shown in FIG. 2 are generated in advance in accordance with what kind of lifestyle pattern will be identified in what kind of lifestyle pattern identification period by the lifestyle rhythm estimation device 1, and are stored in the storage unit 17. Here, lifestyle rhythm estimation device 1 identifies a lifestyle pattern that indicates the time periods during which the subject eats meals in a day, in other words, the time periods during which the subject opens and closes the refrigerator door in a day. Therefore, basic patterns such as those shown in FIG. 2, which are generalized types of the time periods during which the refrigerator door is opened and closed in a day, are generated in advance. FIGS. 2A, 2B, 2C, and 2D each show basic patterns of the time periods during which the refrigerator door is opened and closed in a day (from midnight to midnight). The basic pattern of the timing at which the refrigerator door is opened and closed can be represented, for example, by the waveform of daily behavior-related data acquired from sensor 2. More specifically, Figure 2A shows a basic pattern in which the refrigerator door is opened and closed in the morning and evening hours, Figure 2B shows a basic pattern in which the refrigerator door is opened and closed in the morning, afternoon, and evening hours, Figure 2C shows a basic pattern in which the refrigerator door is opened and closed only during the afternoon hours, and Figure 2D shows a basic pattern in which the refrigerator door is not opened or closed at all in a day. As mentioned above, sensor 2 is assumed to be a sensor that uses magnetism to detect whether a circuit is closed or open, and to take the binary value of 0 or 1. The basic pattern of the time period during which the refrigerator door is opened and closed is represented by the cumulative time the refrigerator door is opened and closed over a certain time interval, based on the sensor values ​​indicated in the daily activity-related data. For example, the time the refrigerator door is open is cumulatively calculated over 30-minute intervals. In this case, if the refrigerator door is open for a cumulative total of 20 minutes out of 30 minutes, the sensor value on the vertical axis will be 20; if the refrigerator door is open for 10 minutes, the sensor value on the vertical axis will be 10; and if the refrigerator door is not opened at all, the sensor value on the vertical axis will be 0. This results in a graph of sensor values ​​that, for example, have peaks during times when the refrigerator door was open for a long period of time (i.e., the refrigerator was used many times).

[0023] For example, before shipping a product, a developer or the like determines a basic pattern, generates basic pattern data indicating the determined basic pattern, and stores this in the storage unit 17. For example, the developer determines a basic pattern based on daily behavior-related data previously acquired from a sensor 2 installed in a refrigerator used by multiple people (hereinafter referred to as "subjects"), and generates basic pattern data indicating the determined basic pattern. It should be noted that this is merely an example, and for example, developers may determine basic patterns based on knowledge or the like, rather than on the daily activity-related data acquired from the sensor 2, and generate basic pattern data. The basic pattern data may be data that indicates a generalized life pattern, i.e., a basic pattern, identified by the life rhythm estimation device 1 based on the daily behavior-related data of the subject during the period for identifying the life pattern obtained from the sensor 2. As described above, there can be various generalized lifestyle patterns, so a plurality of basic patterns are determined in advance, and a plurality of basic pattern data representing the plurality of basic patterns is generated.

[0024] When a developer or the like generates a plurality of basic pattern data, the developer or the like assigns data (hereinafter referred to as "pattern specifying data") indicating which basic pattern each basic pattern data indicates to each basic pattern data, and stores the data in the storage unit 17. The pattern specifying data is, for example, data indicating the name of the basic pattern. For example, if developers etc. determine basic patterns such as those shown in Figures 2A, 2B, 2C, and 2D, they may name Figure 2A a "morning-night pattern" because it is a basic pattern in which the refrigerator door is opened and closed in the morning and evening, and assign pattern specifying data indicating this name to the basic pattern data indicative of the basic pattern of Figure 2A. Also, for example, developers etc. may name Figure 2B a "morning-day-night pattern" because it is a basic pattern in which the refrigerator door is opened and closed in the morning, afternoon, and evening, and assign pattern specifying data indicating this name to the basic pattern data indicative of the basic pattern of Figure 2B. Also, for example, developers etc. may name Figure 2C a "day pattern" because it is a basic pattern in which the refrigerator door is opened and closed only in the afternoon, and assign pattern specifying data indicating this name to the basic pattern data indicative of the basic pattern of Figure 2C. Furthermore, for example, since Figure 2D is a basic pattern in which the refrigerator door is not opened or closed in a day, the developers name it the "no use pattern" and assign pattern identification data indicating this name to the basic pattern data indicating the basic pattern of Figure 2D. Here, the pattern specifying data is assumed to be data indicating the name of a basic pattern, but this is merely an example. The pattern specifying data may be, for example, data indicating an ID for each basic pattern, as long as it is data that can identify the basic pattern indicated by the basic pattern data to which it is assigned.

[0025] 2 shows four basic patterns, but this is merely an example. The number of basic patterns is determined appropriately depending on how many generalized patterns of daily activities are expected to occur during the period for identifying lifestyle patterns, and basic pattern data for the determined number of basic patterns should be generated in advance. In the first embodiment, it is assumed that the storage unit 17 stores four basic pattern data indicating four basic patterns as shown in FIG.

[0026] Returning to the description of the lifestyle pattern identification unit 12. The lifestyle pattern identification unit 12 compares the subject's daily behavior-related data acquired by the data acquisition unit 11 with basic pattern data showing four basic patterns as shown in Figure 2, and identifies the subject's lifestyle pattern for one day. For example, the lifestyle pattern identification unit 12 waits until the data acquisition unit 11 acquires one day's worth of daily activity related data, and when the data acquisition unit 11 acquires one day's worth of daily activity related data, extracts the waveform of the sensor value indicated by the one day's worth of daily activity related data. Note that the lifestyle pattern identification unit 12 may acquire past daily activity related data from the storage unit 17. In the following first embodiment, the daily behavior related data of a subject for a period for identifying a lifestyle pattern, here the daily behavior related data of a subject for one day, is also referred to as "subject daily behavior related data".

[0027] The lifestyle pattern identification unit 12 identifies the subject's daily lifestyle pattern based on which of the basic pattern waveforms the sensor value waveform indicated by the subject's subject daily activity related data matches. In other words, the lifestyle pattern identification unit 12 identifies the subject's daily lifestyle pattern based on which of the basic pattern waveforms the sensor value waveform indicated by the subject's subject daily activity related data resembles. The lifestyle pattern identification unit 12 may use a known method for determining the degree of similarity between data to determine which of the waveforms of each basic pattern the waveform of the sensor value indicated by the target daily activity related data is similar to.

[0028] The lifestyle pattern identification unit 12 identifies a basic pattern waveform similar to the sensor value waveform indicated by the target daily activity-related data by, for example, comparing the time at which a peak appears in the sensor value waveform indicated by the target daily activity-related data with the time at which a peak appears in the basic pattern waveform, thereby identifying the subject's daily lifestyle pattern. If the time at which a peak appears in the sensor value waveform indicated by the target daily activity-related data is similar to the time at which a peak appears in the basic pattern waveform, the lifestyle pattern identification unit 12 determines that the sensor value waveform indicated by the target daily activity-related data and the basic pattern waveform are similar. Here, similarity in appearance time means that the difference between the appearance times is within a predetermined range (hereinafter referred to as a "similar time range"). When multiple peaks appear in the waveform of the sensor value indicated by the target daily activity-related data, the lifestyle pattern identification unit 12, for example, compares the appearance time of each peak with the appearance time of a peak in the waveform of the basic pattern to determine whether the waveform of the sensor value indicated by the target daily activity-related data is similar to the waveform of the basic pattern.

[0029] The lifestyle pattern identification unit 12 may, for example, use a trained model in machine learning (hereinafter referred to as a "machine learning model") to estimate the similarity of the waveform of the sensor value indicated by the subject daily activity-related data to the waveform of each basic pattern, and determine, from the estimated similarity, which basic pattern waveform the sensor value waveform indicated by the subject daily activity-related data corresponds to. The machine learning model is a model that receives the subject daily activity-related data as input and outputs the similarity to each basic pattern, and is generated in advance and stored in the storage unit 17 or the like. The lifestyle pattern identification unit 12, for example, identifies the basic pattern having the largest estimated similarity value and equal to or greater than a preset threshold as the subject's lifestyle pattern.

[0030] The lifestyle pattern identification unit 12 may determine whether the waveform of the sensor value indicated by the subject daily activity-related data is similar to the waveform of the basic pattern by, for example, comparing the shape of a waveform of a specific period (hereinafter referred to as a "partial waveform") among the waveforms of the sensor value indicated by the subject daily activity-related data with the shape of the waveform of the basic pattern. For example, when the waveform of the basic pattern contains a waveform with a similar shape to the partial waveform, the lifestyle pattern identification unit 12 identifies the basic pattern containing the waveform with the similar shape as the lifestyle pattern of the subject.

[0031] In addition, if the lifestyle pattern identification unit 12 determines that the waveform of the sensor value indicated by the target daily behavior-related data does not match any of the waveforms of each basic pattern, it identifies the target person's lifestyle pattern as an ``other pattern'' lifestyle pattern that does not match any of the basic patterns.

[0032] When the lifestyle pattern identification unit 12 identifies the lifestyle pattern of the subject, it outputs data relating to the identified lifestyle pattern of the subject (hereinafter referred to as “lifestyle pattern data”) to the lifestyle pattern compilation unit 13. Furthermore, the lifestyle pattern identification unit 12 stores the lifestyle pattern data in the storage unit 17. The lifestyle pattern data is data in which the date and time when a lifestyle pattern is identified, data indicating the identified lifestyle pattern, for example, pattern identification data, and target daily activity-related data are associated with each other. The lifestyle pattern identification unit 12 may use the pattern identification data assigned to the basic pattern data that was used to identify the subject's lifestyle pattern (in other words, the basic pattern data indicating a basic pattern whose waveform is determined to match the waveform of the sensor value indicated by the subject's target daily activity-related data) as data indicating the identified lifestyle pattern in the lifestyle pattern data.

[0033] Lifestyle pattern counting unit 13 counts up the subject's life pattern for a life rhythm estimation period, here one month, based on the subject's life pattern identified by life pattern identification unit 12 and the life pattern counting conditions. In the first embodiment, the lifestyle pattern tallying condition is a condition that sets the tallying unit and tallying method for the subject's lifestyle patterns during the lifestyle rhythm estimation period. The lifestyle pattern tallying conditions are set in advance by a developer or the like, and data indicating the lifestyle pattern tallying conditions (hereinafter referred to as “lifestyle pattern tallying condition data”) is stored in the storage unit 17. For example, developers may be able to update the lifestyle pattern compilation condition data as needed.

[0034] The lifestyle pattern calculation conditions are set, for example, as the following <Condition (1)>. <Condition (1)> The frequency of occurrence of each lifestyle pattern is calculated using each lifestyle pattern as a calculation unit.

[0035] It is only necessary for the lifestyle pattern counting unit 13 to acquire, from the storage unit 17, lifestyle pattern data for a period for estimating lifestyle rhythms, which in this case is one month. Lifestyle pattern tallying unit 13 tally the lifestyle patterns for each lifestyle rhythm estimation period. When it is time to tally the lifestyle patterns, in other words, when the lifestyle pattern data for the lifestyle rhythm estimation period has been stored in storage unit 17, lifestyle pattern tallying unit 13 acquires the lifestyle pattern data for the most recent past month from storage unit 17.

[0036] If the lifestyle pattern counting condition is set to the above-mentioned <Condition (1)>, the lifestyle pattern counting unit 13 will count, for each lifestyle pattern in the lifestyle pattern data for the past month, the frequency (i.e., the number of days) that the lifestyle pattern appeared in the month.

[0037] FIG. 3 is a diagram for explaining an example of a result of tallying lifestyle patterns tallyed by the lifestyle pattern tallying unit 13 in the first embodiment. The tabulation results shown in Figure 3 are a graphical visualization of the subject's lifestyle patterns over a period for estimating lifestyle rhythms, in this case one month, based on the lifestyle pattern tabulation conditions such as the above-mentioned <Condition (1)>.

[0038] After aggregating the subject's lifestyle patterns for the lifestyle rhythm estimation period, lifestyle pattern aggregation section 13 outputs data indicating the aggregation results (hereinafter referred to as “lifestyle pattern aggregation data”) to lifestyle rhythm estimation section 14. The lifestyle pattern summary data here is data indicating the frequency of occurrence of each lifestyle pattern. More specifically, the lifestyle pattern summary data here is data in which pattern identification data, frequency of occurrence, and data indicating a lifestyle rhythm estimation period are associated with each other.

[0039] Life rhythm estimation unit 14 estimates the life rhythm of the subject from the result of the aggregation of the subject's life patterns aggregated by life pattern aggregation unit 13. The lifestyle rhythm estimation unit 14 may estimate the lifestyle rhythm of the subject according to lifestyle rhythm estimation rules (hereinafter referred to as "lifestyle rhythm estimation rules") that have been appropriately set in advance by a developer or the like. After setting the lifestyle rhythm estimation rules, the developer or the like stores data indicating the lifestyle rhythm estimation rules (hereinafter referred to as "lifestyle rhythm estimation rule data") in the storage unit 17. The lifestyle rhythm estimation unit 14 may estimate the lifestyle rhythm of the subject by referring to the lifestyle rhythm estimation rule data stored in the storage unit 17.

[0040] For example, the lifestyle rhythm estimation rule may be set to "use the aggregation result of lifestyle patterns as the lifestyle rhythm." Note that this is merely one example, and the lifestyle rhythm estimation rule may be set to "use the aggregation result of lifestyle patterns and the result of estimating the trend of the lifestyle patterns based on the trend estimation rule (hereinafter referred to as "trend estimation result") as the lifestyle rhythm," or "use the trend estimation result as the lifestyle rhythm." Appropriate rules may be set for the lifestyle rhythm estimation rule.

[0041] The trend estimation rule is a rule that defines what trends are estimated to exist in the subject's lifestyle pattern based on the lifestyle pattern aggregation results, and is set in advance by a developer, etc. When the developer, etc. sets the trend estimation rule, data indicating the set trend estimation rule (hereinafter referred to as "trend estimation rule data") is stored in the storage unit 17 together with the lifestyle rhythm estimation rule data. For example, the tendency estimation rule may be set as follows: "Extract three lifestyle patterns in descending order of frequency, and estimate that the subject tends to frequently participate in the three extracted lifestyle patterns." In this case, lifestyle rhythm estimation unit 14, for example, creates a list of the frequency of each lifestyle pattern, sorts the list in descending order, and extracts the top three lifestyle patterns. Lifestyle rhythm estimation unit 14 then estimates that the subject tends to frequently participate in the three extracted lifestyle patterns. For example, if the lifestyle pattern aggregation unit 13 aggregates the subject's lifestyle patterns as shown in FIG. 3, lifestyle rhythm estimation unit 14 estimates that the subject tends to frequently participate in the "morning-evening pattern," "morning-day-evening pattern," and "daytime pattern."

[0042] For example, the trend estimation rule may be set to "extract lifestyle patterns whose frequency is equal to or greater than a predetermined threshold (hereinafter referred to as the "frequency determination threshold"), and estimate that the subject tends to have a high frequency of the extracted lifestyle patterns." In this case, lifestyle rhythm estimation unit 14 extracts lifestyle patterns whose frequency is equal to or greater than the frequency determination threshold, and estimates that the subject tends to frequently participate in the extracted lifestyle patterns. For example, suppose the frequency determination threshold is "7 days," and the lifestyle pattern aggregation unit 13 aggregates the subject's lifestyle patterns as shown in FIG. 3, with the frequencies of the "morning-evening pattern," "morning-afternoon-evening pattern," and "afternoon pattern" being 10 days, 9 days, and 8 days, respectively. In this case, lifestyle rhythm estimation unit 14 estimates that the subject tends to frequently participate in the "morning-evening pattern," "morning-afternoon-evening pattern," and "afternoon pattern." The frequency determination threshold is set in advance by a developer or the like along with the tendency estimation rules and stored in storage unit 17.

[0043] In the following first embodiment, as an example, it is assumed that the life rhythm estimation rule is set to "the life pattern aggregation result and the trend estimation result are used as the life rhythm."

[0044] Life rhythm estimation section 14 outputs data relating to the estimated life rhythm of the subject (hereinafter referred to as “life rhythm data”) to change detection section 15. Furthermore, the life rhythm estimation unit 14 stores the life rhythm data in the storage unit 17. The lifestyle rhythm data includes the date and time when the subject's lifestyle rhythm was estimated and data indicating the estimated lifestyle rhythm of the subject. Specifically, the data indicating the subject's estimated lifestyle rhythm is the lifestyle pattern aggregation result of the subject aggregated by lifestyle pattern aggregation unit 13 and data indicating the tendency of the subject's lifestyle pattern estimated by lifestyle rhythm estimation unit 14 from the aggregation result.

[0045] The change detection unit 15 detects a change in the subject's life rhythm estimated by the life rhythm estimation unit 14. More specifically, the change detection unit 15 compares the subject's lifestyle rhythm (referred to as a first lifestyle rhythm) estimated by the lifestyle rhythm estimation unit 14 with another lifestyle rhythm (referred to as a second lifestyle rhythm) of the subject estimated by the lifestyle rhythm estimation unit 14, and detects a change in the subject's lifestyle rhythm. More specifically, change detection unit 15 detects a change in the subject's lifestyle rhythm by comparing a lifestyle pattern aggregation result (hereinafter referred to as a first aggregation result) indicated by the subject's first lifestyle rhythm estimated by lifestyle rhythm estimation unit 14 with a lifestyle pattern aggregation result (hereinafter referred to as a second aggregation result) indicated by the subject's second lifestyle rhythm estimated by lifestyle rhythm estimation unit 14. Note that when lifestyle rhythm estimation unit 14 estimates the trend estimation result as the subject's lifestyle rhythm, change detection unit 15 may acquire, from lifestyle rhythm estimation unit 14, the lifestyle pattern aggregation result that served as the basis for estimating the lifestyle pattern trend, together with lifestyle rhythm data.

[0046] In the first embodiment, for example, the first lifestyle rhythm is assumed to be the most recent lifestyle rhythm estimated by the lifestyle rhythm estimation unit 14, and the second lifestyle rhythm is assumed to be a past lifestyle rhythm estimated by the lifestyle rhythm estimation unit 14 before the first lifestyle rhythm was estimated. The first lifestyle rhythm is assumed to be the lifestyle rhythm of the subject during the first lifestyle rhythm estimation period estimated by the lifestyle rhythm estimation unit 14. The second lifestyle rhythm is assumed to be the lifestyle rhythm of the subject during the second lifestyle rhythm estimation period estimated by the lifestyle rhythm estimation unit 14. It is assumed that the extent to which the second lifestyle rhythm is estimated from lifestyle pattern data of the second lifestyle rhythm estimation period is set in advance. This is merely one example, and the change detection unit 15 may receive data (hereinafter referred to as "comparison designation data") from the subject or the like specifying the extent to which the second lifestyle rhythm is estimated from the lifestyle pattern of the second lifestyle rhythm estimation period, and set the extent to which the second lifestyle rhythm is estimated from the lifestyle pattern of the second lifestyle rhythm estimation period based on the received comparison designation data. The comparison designation data is, for example, data specifying the second lifestyle rhythm estimation period. The subject or the like inputs the comparison designation data from, for example, an input unit (not shown) provided in a mobile terminal or PC. The change detection unit 15 receives the comparison designation data from the mobile terminal or PC. Furthermore, for example, the subject may be able to specify not only the second life rhythm estimation period but also the first life rhythm estimation period. In this case, the change detection unit 15 may receive comparison specification data specifying the first life rhythm estimation period and the second life rhythm estimation period, and determine which life rhythm pattern data of which life rhythm estimation period the first life rhythm and the second life rhythm should be estimated from.

[0047] Here, as an example, the first life rhythm estimation period is the most recent month, and the second life rhythm estimation period is the month six months ago. Change detection section 15 acquires life rhythm data indicating the first life rhythm and life rhythm data indicating the second life rhythm from storage section 17, and compares the two. The change detection unit 15 may compare the first and second lifestyle rhythms of the subject in accordance with lifestyle rhythm comparison rules (hereinafter referred to as "lifestyle rhythm comparison rules") that have been set in advance by a developer or the like. After setting the lifestyle rhythm comparison rules, the developer or the like stores data indicating the lifestyle rhythm comparison rules (hereinafter referred to as "lifestyle rhythm comparison rule data") in the storage unit 17. The change detection unit 15 may compare the lifestyle rhythms of the subject by referring to the lifestyle rhythm comparison rule data stored in the storage unit 17.

[0048] For example, the lifestyle rhythm comparison rule is set to "compare the frequency of each lifestyle pattern, and if there is a change in frequency, it is assumed that there is a change in the lifestyle rhythm." Furthermore, for example, the lifestyle rhythm comparison rule may be set to a rule that determines how a lifestyle rhythm has changed based on the combination of which lifestyle patterns have increased and which have decreased as a result of comparing the composition ratio of the aggregated results of the lifestyle patterns indicated by the first lifestyle rhythm with the composition ratio of the aggregated results of the lifestyle patterns indicated by the second lifestyle rhythm.

[0049] FIG. 4 is a diagram for explaining an example of detection of a change in life rhythm by change detection section 15 in the first embodiment. In Figure 4, the left side of the figure is a graph showing the second lifestyle rhythm estimation period, in other words, the lifestyle rhythm from one month ago (second lifestyle rhythm), and the right side of the figure is a graph showing the first lifestyle rhythm estimation period, in other words, the most recent lifestyle rhythm (first lifestyle rhythm). Figure 4 also shows the results of tallying the lifestyle patterns represented by the first lifestyle rhythm and the second lifestyle rhythm. For example, the lifestyle rhythm comparison rule is set to a rule that, as a result of comparing the component ratio of the aggregated results of lifestyle patterns indicated by a first lifestyle rhythm with the component ratio of the aggregated results of lifestyle patterns indicated by a second lifestyle rhythm, determines how a lifestyle rhythm has changed based on the combination of which lifestyle patterns have increased and which lifestyle patterns have decreased. Specifically, the lifestyle rhythm comparison rule is set to a rule that states, "If the 'morning-afternoon-evening pattern' has decreased and the 'morning-evening pattern' has increased, it is determined that there has been a change such that the number of days on which the refrigerator is not used during the day has increased." In this case, the change detection unit 15 compares, for example, the component ratio of the aggregation results of the lifestyle patterns indicated by the first lifestyle rhythm with the component ratio of the aggregation results of the lifestyle patterns indicated by the second lifestyle rhythm, in this case, the frequency of each lifestyle pattern in the first lifestyle rhythm with the frequency of each lifestyle pattern in the corresponding second lifestyle rhythm, and detects a change in lifestyle rhythm that "the number of days on which the refrigerator is not used during the day has increased" because the "morning-afternoon-evening pattern" has decreased and the "morning-evening pattern" has increased.

[0050] Change detection unit 15 outputs data indicating the results of detecting changes in life rhythms (hereinafter referred to as “change detection result data”) to data provision unit 16. The change detection unit 15 may store the change detection result data in the storage unit 17. The change detection result data is, for example, data indicating the date and time when a change in lifestyle rhythm was detected, data indicating whether or not there was a change in lifestyle rhythm, and data indicating what kind of change was detected if a change in lifestyle rhythm was detected, which are associated with each other. The change detection result data may further include associated lifestyle rhythm data indicating a first lifestyle rhythm and lifestyle rhythm data indicating a second lifestyle rhythm.

[0051] The data providing unit 16 outputs data relating to the subject's life rhythm estimated by the life rhythm estimating unit 14 (hereinafter referred to as "provided data").

[0052] The data providing unit 16 generates data explaining the details of the subject's life rhythm based on, for example, the life rhythm data output from the life rhythm estimating unit 14, and outputs the generated data to the output device 3 as provided data.

[0053] For example, if the output device 3 is a display device, the provided data is data for displaying the content of the subject's life rhythm in the form of a graph, a diagram, a message, or the like. The data providing unit 16 causes the output device 3 to display a graph, a diagram, a message, or the like that explains the details of the subject's life rhythm. For example, the data providing unit 16 displays the tabulation results for each of a plurality of lifestyle patterns indicated by life rhythms in a graph format (see, for example, FIGS. 3 and 4). For example, the data providing unit 16 may display a message explaining the tendency of the lifestyle patterns in the lifestyle rhythm, such as "The most common lifestyle pattern is the 'morning-evening pattern', followed by the 'daytime pattern' and the 'morning-daytime-evening pattern'." The data providing unit 16 may generate provision data for displaying the message from data indicating the tendency of the subject's lifestyle pattern, which is included in the lifestyle rhythm data.

[0054] Furthermore, the data providing unit 16 may output to the output device 3, for example, data proposing measures to improve the lifestyle rhythm based on the lifestyle rhythm data output from the lifestyle rhythm estimating unit 14, as provided data.

[0055] For example, the data providing unit 16 estimates improvement measures for the subject's lifestyle rhythm according to a preset rule for determining improvement measures for lifestyle rhythm (hereinafter referred to as "improvement measure determination rule"). The improvement measure determination rule is a rule that defines what kind of improvement measure should be proposed depending on what kind of lifestyle rhythm the subject has. The improvement measure determination rules are set in advance by developers, etc. The developers, etc. set the improvement measure determination rules in advance, for example, from a collection of past cases showing what kind of improvement measures should be taken for what kind of lifestyle rhythms to improve physical and mental health. After setting the improvement measure determination rules, the developers, etc. store data indicating the improvement measure determination rules (hereinafter referred to as "improvement measure determination rule data") in the memory unit 17. The data providing unit 16 can identify the improvement measure determination rules from the improvement measure determination rule data stored in the memory unit 17. Note that in FIG. 1, the arrow connecting the data providing unit 16 and the memory unit 17 is omitted.

[0056] Furthermore, for example, the data providing unit 16 may use LLMs (Large Language Models) to determine improvement measures for the subject's lifestyle rhythm. The data providing unit 16 inputs the lifestyle rhythm data output from the lifestyle rhythm estimation unit 14 into the LLMs to obtain data indicating improvement measures for the subject's lifestyle rhythm. As described above, the lifestyle rhythm estimation unit 14 may estimate the lifestyle pattern tendency from the result of the aggregation of the subject's lifestyle patterns aggregated by the lifestyle pattern aggregation unit 13 in accordance with the lifestyle rhythm estimation rules, but for example, the data providing unit 16 may input the lifestyle rhythm data output from the lifestyle rhythm estimation unit 14 into the LLM to estimate the lifestyle pattern tendency of the subject. For example, the lifestyle rhythm estimation unit 14 may estimate the lifestyle pattern tendency of the subject using the LLM. Furthermore, for example, in detecting a change in the subject's lifestyle rhythm by the change detection unit 15 described above, the change detection unit 15 may input lifestyle rhythm data indicating a first lifestyle rhythm and lifestyle rhythm data indicating a second lifestyle rhythm to the LLM, rather than using a method based on lifestyle rhythm comparison rules, and have the LLM determine or explain whether or how the subject's lifestyle rhythm has changed. The data providing unit 16 may then output, as provided data, change detection result data indicating whether or how the subject's lifestyle rhythm has changed, which the change detection unit 15 has caused the LLM to determine or explain.

[0057] For example, if the output device 3 is a display device, the data providing unit 16 outputs provided data that suggests an improvement measure for the subject's lifestyle rhythm to the output device 3, and causes the output device 3 to display the improvement measure. The output format in which the data providing unit 16 generates the provided data that causes the output device 3 to output measures to improve the life rhythm of the subject is determined in advance by the developer or the like.

[0058] Furthermore, when change detection result data is output from the change detection unit 15, the data providing unit 16 outputs the change detection result data to the output device 3 as provided data. For example, if the output device 3 is a display device, the data providing unit 16 causes the output device 3 to display a message, icon, etc. indicating whether or not there is a change in the lifestyle rhythm and a message, icon, etc. indicating what kind of change has been detected, based on the change detection result data. The message, icon, etc. indicating what kind of change has been detected may be, for example, a message, icon, etc. calling attention to a disruption in the lifestyle rhythm, or a message, icon, etc. praising an improvement in the lifestyle rhythm or the ability to maintain a good lifestyle rhythm. Note that this is merely an example, and the output format in which the data providing unit 16 generates the provided data that causes the output device 3 to output the change detection result is determined in advance by a developer or the like.

[0059] The operation of the lifestyle rhythm estimation device 1 according to the first embodiment will be described. FIG. 5 is a flowchart for explaining the operation of the lifestyle pattern identification process in the lifestyle rhythm estimation device 1 according to the first embodiment. FIG. 6 is a flowchart for explaining the operation of the lifestyle rhythm estimation process in the lifestyle rhythm estimation device 1 according to the first embodiment.

[0060] The operation of the lifestyle pattern identification process shown in the flowchart of FIG. 5 starts, for example, when the lifestyle rhythm estimation device 1 is powered on, and is repeated until the lifestyle rhythm estimation device 1 is powered off. The operation of the lifestyle pattern identification process shown in the flowchart of FIG. 6 begins, for example, when a subject or the like inputs an instruction to start the operation and is repeated until the subject or the like inputs an instruction to stop the operation. For example, the subject or the like inputs a start instruction to start estimating a lifestyle rhythm from a mobile device or PC. Upon receiving the start instruction, a control unit (not shown) of the lifestyle rhythm estimation device 1 outputs control signals to the lifestyle pattern compilation unit 13, the lifestyle rhythm estimation unit 14, the change detection unit 15, and the data providing unit 16 to start the operation. Furthermore, for example, the subject or the like inputs an end instruction to stop estimating the lifestyle rhythm from a mobile device or PC. Upon receiving the end instruction, the control unit (not shown) of the lifestyle rhythm estimation device 1 outputs control signals to the lifestyle pattern compilation unit 13, the lifestyle rhythm estimation unit 14, the change detection unit 15, and the data providing unit 16 to stop the operation. The operation of the lifestyle pattern identification process shown in the flowchart of FIG. 5 and the operation of the lifestyle rhythm estimation process shown in the flowchart of FIG. 6 are performed in parallel.

[0061] First, the operation of the lifestyle rhythm estimation device 1 in the lifestyle pattern identification process shown in the flowchart of FIG. 5 will be described.

[0062] The data acquisition unit 11 acquires daily activity related data from the sensor 2 (step ST1). The data acquiring unit 11 outputs the acquired daily activity related data to the lifestyle pattern identifying unit 12. The data acquiring unit 11 also causes the storage unit 17 to store the acquired daily activity related data.

[0063] The lifestyle pattern identification unit 12 compares the daily activity related data acquired by the data acquisition unit 11 in step ST1 with the basic pattern data to identify the subject's lifestyle pattern identification period, in this case, the subject's lifestyle pattern for one day (step ST2). When lifestyle pattern identification unit 12 identifies the lifestyle pattern of the subject, it outputs the lifestyle pattern data of the subject to lifestyle pattern compilation unit 13. Furthermore, lifestyle pattern identification unit 12 causes storage unit 17 to store the lifestyle pattern data.

[0064] Next, the operation of the lifestyle rhythm estimation device 1 in the lifestyle rhythm estimation process shown in the flowchart of FIG. 6 will be described.

[0065] Lifestyle pattern counting section 13 acquires the subject's lifestyle pattern data for a lifestyle rhythm estimation period, here one month, from lifestyle pattern identification section 12 or storage section 17 (step ST11). If the subject's lifestyle pattern data for the most recent month is not stored in the memory unit 17, the lifestyle pattern counting unit 13 waits to execute the processing of step ST11 until the subject's lifestyle pattern data for the most recent month is stored in the memory unit 17, and executes the processing of step ST11 when the subject's lifestyle pattern data for the most recent month is stored in the memory unit 17.

[0066] The life pattern counting unit 13 counts up the life patterns of the subject during the life rhythm estimation period based on the life pattern data of the subject acquired in step ST11 and the life pattern counting conditions (step ST12). After aggregating the lifestyle patterns of the subject, the lifestyle pattern aggregation unit 13 outputs the lifestyle pattern aggregation data to the lifestyle rhythm estimation unit 14.

[0067] The life rhythm estimation unit 14 estimates the life rhythm of the subject from the result of the collection of the life patterns of the subject collected by the life pattern collection unit 13 in step ST12 (step ST13). Life rhythm estimation section 14 outputs life rhythm data relating to the estimated life rhythm of the subject to change detection section 15. Furthermore, the life rhythm estimation unit 14 stores the life rhythm data in the storage unit 17.

[0068] The change detection unit 15 detects a change in the subject's life rhythm estimated by the life rhythm estimation unit 14 in step ST13 (step ST14). Change detection unit 15 outputs change detection result data indicating the result of detecting a change in life rhythm to data provision unit 16. The change detection unit 15 may store the change detection result data in the storage unit 17.

[0069] The data providing unit 16 outputs provided data relating to the subject's life rhythm estimated by the life rhythm estimating unit 14 in step ST14 (step ST15).

[0070] In this way, lifestyle rhythm estimation device 1 acquires daily activity-related data related to the subject's daily activities detected by sensor 2, compares the acquired daily activity-related data with multiple basic pattern data each indicating a different basic lifestyle pattern, and identifies the subject's lifestyle pattern for a period for identifying a lifestyle pattern, here one day. Based on the identified subject's lifestyle pattern and lifestyle pattern aggregation conditions, lifestyle rhythm estimation device 1 aggregates the subject's lifestyle pattern for a lifestyle rhythm estimation period, here one month, and estimates the subject's lifestyle rhythm from the aggregation results of the aggregated subject's lifestyle pattern. As a result, the lifestyle rhythm estimation device 1 can estimate a generalized lifestyle rhythm by estimating the subject's lifestyle rhythm during the lifestyle rhythm estimation period, taking into account that people do not always follow a fixed lifestyle pattern during each of the multiple lifestyle pattern identification periods but rather live with a variety of lifestyle patterns. In other words, the lifestyle rhythm estimation device 1 can generalize and grasp the subject's lifestyle rhythm.

[0071] In the above-described first embodiment, as an example, a condition such as <Condition (1)> is set as the lifestyle pattern compilation condition, but this is merely an example. The lifestyle pattern compilation condition may be set to a condition other than <Condition (1)>. For example, the lifestyle pattern counting condition may be set as follows: <Condition (2)>. <Condition (2)> The percentage of lifestyle patterns for each day of the week is calculated using the day of the week as the unit of calculation.

[0072] If the lifestyle pattern tallying condition is set to the above-mentioned <Condition (2)>, the lifestyle pattern tallying unit 13 will tally the proportion of lifestyle patterns for each day of the week for the lifestyle pattern data within the past month.

[0073] FIG. 7 is a diagram illustrating another example of the result of tallying the lifestyle patterns tallyed by the lifestyle pattern tallying unit 13 in the first embodiment. The counting results shown in FIG. 7 are visualized in a graph format based on the lifestyle pattern counting conditions such as the above-mentioned <Condition (2)>.

[0074] Lifestyle pattern tallying unit 13 tallys up the subject's lifestyle patterns over a lifestyle rhythm estimation period, in this case one month, and outputs lifestyle pattern tally data indicating the tallying results to lifestyle rhythm estimation unit 14. In this case, the lifestyle pattern tally data is data indicating the proportion of lifestyle patterns for each day of the week. More specifically, in this case, the lifestyle pattern tally data is data in which the day of the week, the proportion of each lifestyle pattern, and data indicating the lifestyle rhythm estimation period are associated with each other.

[0075] In this way, the lifestyle pattern counting unit 13 can also count the lifestyle patterns of the subject on a day-by-day basis during the lifestyle rhythm estimation period. Note that the above-mentioned <Condition (2)> is merely an example. For example, the lifestyle pattern tallying condition may be set to "calculate the frequency of occurrence of lifestyle patterns for each day of the week, with each day of the week being used as a tallying unit."

[0076] In this case, the trend estimation rule may be set to, for example, "extract a lifestyle pattern with a high proportion for each day of the week, and estimate that the subject tends to have a high proportion of the extracted lifestyle pattern on each day of the week." Alternatively, the trend estimation rule may be set to, for example, "extract a lifestyle pattern with a proportion equal to or greater than a preset threshold (hereinafter referred to as a "proportion determination threshold") for each day of the week, and estimate that the subject tends to have a high proportion of the extracted lifestyle pattern on each day of the week." For example, if the result of the aggregation of the subject's lifestyle patterns aggregated by the lifestyle pattern aggregation unit 13 is as shown in FIG. 7, the lifestyle rhythm estimation unit 14 estimates that the subject tends to have more "morning-evening pattern" on Mondays and Tuesdays, more "morning-afternoon-evening pattern" on Wednesdays, Thursdays, and Fridays, and more "afternoon pattern" on Saturdays and Sundays.

[0077] In this case as well, the change detection unit 15 may compare the first life rhythm and the second life rhythm of the subject in accordance with the life rhythm comparison rule. The lifestyle rhythm comparison rule may, for example, be set to "compare the proportion of each lifestyle pattern for each day of the week, and if there is a change in the proportion of each lifestyle pattern for each day of the week, it is determined that there is a change in the lifestyle rhythm." The lifestyle rhythm comparison rule may, for example, be set to "compare the frequency of occurrence of each lifestyle pattern for each day of the week, and if there is a change in the frequency of occurrence of each lifestyle pattern for each day of the week, it is determined that there is a change in the lifestyle rhythm." Furthermore, for example, the lifestyle rhythm comparison rule may be set to determine how a lifestyle rhythm has changed based on the combination of which lifestyle patterns have increased in proportion and which lifestyle patterns have decreased in proportion for each day of the week, as a result of comparing the composition ratio of the aggregated results of the lifestyle patterns indicated by the first lifestyle rhythm with the composition ratio of the aggregated results of the lifestyle patterns indicated by the second lifestyle rhythm.

[0078] FIG. 8 is a diagram for explaining another example of detection of a change in life rhythm by change detection section 15 in the first embodiment. In Figure 8, the left diagram is a graph visualizing the second lifestyle rhythm estimation period, in other words, the lifestyle rhythm from one month ago (second lifestyle rhythm), and the right diagram is a graph visualizing the first lifestyle rhythm estimation period, in other words, the most recent lifestyle rhythm (first lifestyle rhythm). Note that Figure 8 shows the aggregation results of the lifestyle patterns represented by the first lifestyle rhythm and the second lifestyle rhythm. For example, the lifestyle rhythm comparison rule may be set to compare the component ratios of the aggregated results of lifestyle patterns indicated by a first lifestyle rhythm with the component ratios of the aggregated results of lifestyle patterns indicated by a second lifestyle rhythm, and determine how lifestyle rhythms are changed based on the combination of which lifestyle pattern percentages have increased and which lifestyle pattern percentages have decreased for each day of the week. Specifically, the lifestyle rhythm comparison rule may be set to state that "if the 'morning-afternoon-evening pattern' on a certain day of the week has decreased and the 'morning-evening pattern' has increased, it is determined that there has been a change such that the number of days on which the refrigerator is not used during the day on that certain day of the week has increased."

[0079] In this case, the change detection unit 15 compares, for example, the component ratio of the aggregation results of the lifestyle patterns indicated by the first lifestyle rhythm with the component ratio of the aggregation results of the lifestyle patterns indicated by the second lifestyle rhythm, in this case, the proportion of each lifestyle pattern for each day of the week in the first lifestyle rhythm with the proportion of each lifestyle pattern for each day of the week in the corresponding second lifestyle rhythm, and detects a change in lifestyle rhythm that "the number of days on Wednesdays when the refrigerator is not used during the day has increased" because the "morning-afternoon-evening pattern" on Wednesdays has decreased and the "morning-evening pattern" has increased.

[0080] Furthermore, in the above-described first embodiment, the period for identifying a lifestyle pattern is set to one day and the period for estimating a lifestyle rhythm is set to one month as an example, but this is merely an example. For example, the period for identifying a lifestyle pattern may be a set time period such as morning (9:00 to 12:00), or may be a period consisting of multiple days such as one week. Furthermore, for example, the life rhythm estimation period may be one year, or may be a period including one or more cycles, each cycle consisting of a number of days. A specific example will be explained below. In the following example, the lifestyle rhythm estimation period is assumed to be one month, including multiple eight-day cycles. The lifestyle pattern identification period is assumed to be one day. In this case, for example, the subject or the like may specify, via a mobile device or PC, which day is the first day and how many days are to be counted as one cycle, and a control unit (not shown) in lifestyle rhythm estimation device 1 may accept this and instruct lifestyle pattern identification unit 12 on the timing to start counting the cycle according to the specification from the subject or the like. Also, for example, lifestyle pattern identification unit 12 may determine the number of days that constitute one cycle from the periodicity of the daily activity-related data stored in storage unit 17. Note that lifestyle pattern identification unit 12 may determine the periodicity of the daily activity-related data using a known technique for determining periodicity.

[0081] Furthermore, it is assumed that the lifestyle pattern tallying condition is set to, for example, the following <Condition (3)>. <Condition (3)> The frequency or proportion of occurrence of daily lifestyle patterns is calculated using each day in the cycle as a calculation unit.

[0082] If the lifestyle pattern tallying condition is set to the above-mentioned <Condition (3)>, the lifestyle pattern tallying unit 13 will tally the proportion of daily lifestyle patterns in the cycle for the lifestyle pattern data within the past month.

[0083] FIG. 9 is a diagram illustrating another example of the result of tallying the lifestyle patterns tallyed by the lifestyle pattern tallying unit 13 in the first embodiment. The counting results shown in FIG. 9 are visualized in a graph format based on the lifestyle pattern counting condition such as the above-mentioned <Condition (3)>.

[0084] Lifestyle pattern tallying unit 13 tallys up the subject's lifestyle patterns for a lifestyle rhythm estimation period, in this case one month including multiple 8-day cycles, and outputs lifestyle pattern tally data indicating the tallying results to lifestyle rhythm estimation unit 14. In this case, the lifestyle pattern tally data is data indicating the proportion of lifestyle patterns for each day in the cycle. More specifically, the lifestyle pattern tally data in this case is data in which data indicating the day in the cycle (which day it is) is associated with the proportion of each lifestyle pattern and data indicating the lifestyle rhythm estimation period.

[0085] In this way, the lifestyle pattern counting unit 13 can also count the lifestyle patterns of the subject on a daily cycle basis during the lifestyle rhythm estimation period. Note that the above-mentioned <Condition (3)> is merely an example. For example, the lifestyle pattern tallying condition may be set to "tally up the frequency of occurrence of lifestyle patterns for each day, with each day in the cycle being the tallying unit."

[0086] In this case, the trend estimation rule may be set to, for example, "extract a lifestyle pattern with a high proportion for each day in the cycle, and estimate that the subject tends to have a high proportion of the extracted lifestyle pattern for each day." Alternatively, the trend estimation rule may be set to, for example, "extract a lifestyle pattern with a proportion equal to or greater than a predetermined proportion determination threshold for each day in the cycle, and estimate that the subject tends to have a high proportion of the extracted lifestyle pattern for each day in the cycle." For example, if the result of the aggregation of the subject's lifestyle patterns aggregated by lifestyle pattern aggregation unit 13 is as shown in FIG. 9, lifestyle rhythm estimation unit 14 estimates that the subject tends to have more "morning-evening pattern" on the first and second days of the cycle, more "day-evening pattern" on the third and fourth days, and more "morning-day-evening pattern" from the fifth day onwards.

[0087] In this case as well, the change detection unit 15 may compare the first life rhythm and the second life rhythm of the subject in accordance with the life rhythm comparison rule. The lifestyle rhythm comparison rule may, for example, be set to "compare the proportion of each lifestyle pattern for each day in the cycle, and if there is a change in the proportion of each lifestyle pattern for each day, it is determined that there is a change in the lifestyle rhythm." The lifestyle rhythm comparison rule may, for example, be set to "compare the frequency of occurrence of each lifestyle pattern for each day in the cycle, and if there is a change in the frequency of occurrence of each lifestyle pattern for each day, it is determined that there is a change in the lifestyle rhythm." Furthermore, for example, the lifestyle rhythm comparison rule may include a rule for detecting a change in lifestyle rhythm based on the combination of which lifestyle patterns have increased in proportion and which lifestyle patterns have decreased in proportion for each day in the cycle, as a result of comparing the composition ratio of the aggregated results of lifestyle patterns represented by the first lifestyle rhythm with the composition ratio of the aggregated results of lifestyle patterns represented by the second lifestyle rhythm.

[0088] FIG. 10 is a diagram for explaining another example of detection of a change in life rhythm by change detection section 15 in the first embodiment. In Figure 10, the left diagram is a graph visualizing the second lifestyle rhythm estimation period, in other words, the lifestyle rhythm from one month ago (second lifestyle rhythm), and the right diagram is a graph visualizing the first lifestyle rhythm estimation period, in other words, the most recent lifestyle rhythm (first lifestyle rhythm). Note that Figure 10 shows the aggregation results of lifestyle patterns represented by the first lifestyle rhythm and the second lifestyle rhythm. For example, the lifestyle rhythm comparison rule may be set to compare the component ratios of the aggregated results of lifestyle patterns represented by a first lifestyle rhythm with the component ratios of the aggregated results of lifestyle patterns represented by a second lifestyle rhythm, and determine how a lifestyle rhythm has changed based on the combination of which lifestyle pattern's percentage has increased and which lifestyle pattern's percentage has decreased for each day of the week. Specifically, the lifestyle rhythm comparison rule may be set to state that "if the 'morning-day-night pattern' on the nth day in the cycle decreases and the 'day-night pattern' increases, it is determined that there has been a change such that the number of days on which the refrigerator is not used on the morning of the nth day increases."

[0089] In this case, the change detection unit 15 compares, for example, the component ratio of the aggregation results of the lifestyle patterns indicated by the first lifestyle rhythm with the component ratio of the aggregation results of the lifestyle patterns indicated by the second lifestyle rhythm, in this case, the proportion of each lifestyle pattern for each day in the cycle of the first lifestyle rhythm with the proportion of each lifestyle pattern for each day in the cycle of the corresponding second lifestyle rhythm, and detects a change in lifestyle rhythm, namely, "the refrigerator is no longer used in the mornings after the fifth day," since the "morning-day-night pattern" decreases from the fifth to eighth days in the cycle and the "day-night pattern" increases. For example, if the subject specifies that the cycle should be counted as one cycle of eight days, in accordance with the subject's own eight-day work cycle of four days on and four days off, the detection result of the change in lifestyle rhythm as described above by the change detection unit 15 can be said to indicate a change in lifestyle rhythm in that the subject has stopped using the refrigerator in the mornings on his or her days off.

[0090] In addition, in the above-described embodiment 1, the change detection unit 15 detects a change in the lifestyle rhythm during the two lifestyle rhythm estimation periods (the first lifestyle rhythm estimation period and the second lifestyle rhythm estimation period) by comparing the lifestyle rhythms (the first lifestyle rhythm estimation period and the second lifestyle rhythm estimation period) during the two lifestyle rhythm estimation periods, but this is merely one example. In the above-described first embodiment, the change detection unit 15 may detect a change in the subject's lifestyle rhythm based on the ratio of each lifestyle pattern in the aggregated results of a plurality of lifestyle patterns, which are indicated by the subject's lifestyle rhythm estimated over a plurality of time-series lifestyle rhythm estimation periods. More specifically, the change detection unit 15 may detect a change in the subject's lifestyle rhythm from, for example, a transition in the ratio of each lifestyle pattern in the aggregated results of multiple lifestyle patterns, which is indicated by the subject's lifestyle rhythm estimated over multiple time-series lifestyle rhythm estimation periods, the presence or absence of a sudden change in the ratio, or a change in the variance of the ratio.

[0091] FIG. 11 is a diagram illustrating an example of a method for detecting a change in a subject's lifestyle rhythm based on the ratio of each lifestyle pattern in the aggregation results of multiple lifestyle patterns, which is indicated by the subject's lifestyle rhythm estimated by the change detection unit 15 over multiple time-series lifestyle rhythm estimation periods in embodiment 1. Figure 11A is a diagram illustrating an example of a method by which the change detection unit 15 detects a change in the subject's lifestyle rhythm from a transition in the ratio of each lifestyle pattern, Figure 11B is a diagram illustrating an example of a method by which the change detection unit 15 detects a change in the subject's lifestyle rhythm from a sudden change in the ratio of each lifestyle pattern, and Figure 11C is a diagram illustrating an example of a method by which the change detection unit 15 detects a change in the subject's lifestyle rhythm by taking into account the ratio and variation of each lifestyle pattern. For convenience, Figure 11 uses "Pattern A," "Pattern B," and "Pattern C," but these "Pattern A," "Pattern B," and "Pattern C" are intended to represent, for example, a "morning-night pattern," a "morning-afternoon-afternoon pattern," and a "afternoon pattern," respectively. In addition, in FIG. 11, the period for identifying a lifestyle pattern is one day, the period for estimating a lifestyle rhythm is one month, and the lifestyle pattern aggregation condition is set to the above <Condition (1)>. The change detection unit 15 detects changes in the subject's lifestyle rhythm from the ratio of each lifestyle pattern in the aggregated results of multiple lifestyle patterns, which are indicated by the subject's lifestyle rhythm estimated over the three lifestyle rhythm estimation periods of ``the month before last,'' ``last month,'' and ``this month.''

[0092] For example, suppose that the ratios of each lifestyle pattern within a lifestyle rhythm estimation period (here, one month) for the month before last, last month, and this month are as shown in Fig. 11A. In this case, the ratios of each lifestyle pattern change significantly from last month to this month. Therefore, change detection unit 15 detects that the subject's lifestyle rhythm has changed. For example, the change detection unit 15 may predict the ratio of each lifestyle pattern for this month from the ratios for the month before last and last month, and then determine whether the ratio of each lifestyle pattern is changing, thereby detecting whether the subject's lifestyle rhythm has changed.

[0093] Also, for example, suppose that the ratios of each lifestyle pattern within a lifestyle rhythm estimation period (here, one month) for the month before last, last month, and this month are as shown in Fig. 11B. In this case, the ratios of each lifestyle pattern change drastically from last month to this month. Therefore, the change detection unit 15 detects that the subject's lifestyle rhythm has changed.

[0094] For example, suppose the ratios of each lifestyle pattern within a lifestyle rhythm estimation period (here, one month) for the month before last, last month, and this month are as shown in Fig. 11C. In this case, when looking at the ratios of each lifestyle pattern for the month before last, last month, and this month, there is not much change in themselves, but the variation is large. Therefore, the change detection unit 15 detects that the subject's lifestyle rhythm has changed.

[0095] Furthermore, in the above-described first embodiment, the change detection unit 15 may detect a change in the subject's lifestyle rhythm based on whether or not a new lifestyle pattern has emerged in the subject's lifestyle rhythm estimated by the lifestyle rhythm estimation unit 14. When a new lifestyle pattern has emerged in the subject's lifestyle rhythm, the change detection unit 15 detects that there has been a change in the subject's lifestyle rhythm. The change detection unit 15 may, for example, acquire past lifestyle rhythm data stored in the memory unit 17 by the lifestyle rhythm estimation unit 14, and compare the multiple lifestyle patterns indicated by the past lifestyle rhythm with the multiple lifestyle patterns indicated by the latest lifestyle rhythm to determine whether a new lifestyle pattern has emerged in the subject's lifestyle rhythm.

[0096] Furthermore, for example, when the life rhythm estimation period is a period including one or more cycles each consisting of a plurality of days (see FIG. 9), the change detection unit 15 may detect a change in the life rhythm of the subject depending on whether or not there has been a change in the periodicity of the daily activity related data, based on the daily activity related data acquired by the data acquisition unit 11. When there has been a change in the periodicity of the daily activity related data, the change detection unit 15 detects that there has been a change in the life rhythm of the subject. The change detection unit 15 may, for example, acquire past daily activity related data that the data acquisition unit 11 has stored in the storage unit 17, and determine whether the periodicity of the daily activity related data has changed.

[0097] In the first embodiment, the data providing unit 16 may output various types of data as provided data in addition to the provided data described above.

[0098] For example, the data providing unit 16 may compare the subject's lifestyle rhythm estimated by the lifestyle rhythm estimation unit 14 with the lifestyle rhythms of other people to extract characteristics of the subject's lifestyle rhythm, and output data explaining the extracted characteristics as provided data. The other people may be friends, etc. For example, the data providing unit 16 compares multiple lifestyle patterns indicated by the subject's lifestyle rhythm with multiple lifestyle patterns indicated by the lifestyle rhythms of other people, and if the subject's lifestyle rhythm contains a lifestyle pattern that is not present in the lifestyle rhythms of other people, extracts pattern-specific data of the lifestyle pattern as a feature of the subject's lifestyle rhythm.The data providing unit 16 then outputs the extracted pattern-specific data to the output device 3 as provision data.

[0099] Furthermore, for example, when the data providing unit 16 outputs data explaining the content of the subject's lifestyle rhythm or data proposing measures to improve the subject's lifestyle rhythm as provided data, it may search for lifestyle rhythms of others that have been estimated in the past and are similar to the subject's lifestyle rhythm, and the data explaining the content of the lifestyle rhythm or data proposing measures to improve the lifestyle rhythm, generated based on the similar lifestyle rhythms found, may be used as the provided data for the subject. For example, it is assumed that previously estimated lifestyle rhythm data of other people and the provided data are stored in the storage unit 17, and the data providing unit 16 may search for lifestyle rhythms of other people that are similar to the lifestyle rhythm of the subject by referring to the storage unit 17. Furthermore, data (hereinafter referred to as "similar data search data") in which lifestyle rhythm data of multiple people, data explaining the lifestyle rhythms indicated by the lifestyle rhythm data, and data proposing measures to improve the lifestyle rhythms indicated by the lifestyle rhythm data are associated with each other may be generated in advance by a developer or the like and stored in the storage unit 17. The data providing unit 16 may search for lifestyle rhythms of other people that are similar to the lifestyle rhythm of the subject by referring to the similar data search data stored in the storage unit 17.

[0100] Furthermore, for example, the data providing unit 16 may compare the subject's lifestyle rhythm estimated by the lifestyle rhythm estimation unit 14 with a preset lifestyle rhythm (hereinafter referred to as the "recommended lifestyle rhythm"), and output an index showing the subject's level of health as provided data. The recommended lifestyle rhythm is the lifestyle rhythm of a healthy person. For example, a developer or the like may prepare lifestyle rhythm data of a healthy person to be used as a benchmark (hereinafter referred to as "recommended lifestyle rhythm data") in advance and store the recommended lifestyle rhythm data in the storage unit 17. The data providing unit 16 calculates the degree of similarity between the subject's lifestyle rhythm and the recommended lifestyle rhythm based on the subject's lifestyle rhythm data output from the lifestyle rhythm estimation unit 14 and the recommended lifestyle rhythm data stored in the storage unit 17. The data providing unit 16 may calculate the similarity based on, for example, the degree of similarity between the lifestyle patterns. The data providing unit 16 sets an index indicating the subject's health level based on the calculated similarity. The data providing unit 16 sets an index indicating that the greater the calculated similarity, in other words, the more similar the subject's lifestyle rhythm is to the recommended lifestyle rhythm, the higher the subject's health level. For example, the data providing unit 16 may rank the subject's health level. Furthermore, for example, the data providing unit 16 may include data indicating whether the subject's lifestyle rhythm is stable, estimated from the variance of the subject's lifestyle patterns over multiple periods, in the index indicating the subject's health level. For example, if the variance of the subject's lifestyle patterns over multiple periods is small, it can be said that the subject's lifestyle rhythm is stable (see, for example, FIG. 11C ).

[0101] In the first embodiment, the lifestyle rhythm estimation device 1 includes the change detection unit 15 and the data provision unit 16, but this is merely an example. The lifestyle rhythm estimation device 1 may be configured without the change detection unit 15 or the data provision unit 16. When the lifestyle rhythm estimation device 1 is configured without the change detection section 15, the operation of the lifestyle rhythm estimation device 1 can omit the processing of step ST14 in the operation described using the flowchart of FIG. When the lifestyle rhythm estimation device 1 is configured without the data providing unit 16, the operation of the lifestyle rhythm estimation device 1 can omit the processing of step ST15 in the operation described using the flowchart of FIG.

[0102] In the first embodiment, the subject's daily behavior, which is the subject of estimation of the life rhythm, is assumed to be eating, but this is merely an example. The subject's daily behavior includes various daily behaviors other than eating, such as sleeping, working, exercise such as walking or running, bathing, or resting. The life rhythm estimation device 1 can estimate the life rhythm of the subject for various daily activities from various daily activity related data detected by the sensor 2 that can be considered to represent various daily activities of the subject.

[0103] 12A and 12B are diagrams illustrating an example of the hardware configuration of the lifestyle rhythm estimation device 1 according to the first embodiment. In the first embodiment, the functions of data acquisition unit 11, lifestyle pattern identification unit 12, lifestyle pattern aggregation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, data provision unit 16, and a control unit (not shown) are realized by processing circuit 101. That is, lifestyle rhythm estimation device 1 includes processing circuit 101 for controlling estimation of the lifestyle rhythm so that the lifestyle rhythm of a subject can be generalized and understood. The processing circuit 101 may be dedicated hardware as shown in FIG. 12A, or may be a processor 104 that executes a program stored in a memory 105 as shown in FIG. 12B.

[0104] When the processing circuitry 101 is dedicated hardware, the processing circuitry 101 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0105] When the processing circuit is processor 104, the functions of data acquisition unit 11, lifestyle pattern identification unit 12, lifestyle pattern compilation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, data provision unit 16, and a control unit (not shown) are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in memory 105. Processor 104 reads and executes the program stored in memory 105 to perform the functions of data acquisition unit 11, lifestyle pattern identification unit 12, lifestyle pattern compilation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, data provision unit 16, and a control unit (not shown). In other words, lifestyle rhythm estimation device 1 includes memory 105 for storing a program that, when executed by processor 104, results in the execution of steps ST1 to ST2 of FIG. 5 or steps ST11 to ST15 of FIG. 6. Furthermore, the programs stored in memory 105 can be said to cause the computer to execute the processing procedures or methods of data acquisition unit 11, lifestyle pattern identification unit 12, lifestyle pattern compilation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, data provision unit 16, and a control unit (not shown). Here, memory 105 corresponds to, for example, non-volatile or volatile semiconductor memory such as RAM, ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), magnetic disk, flexible disk, optical disk, compact disk, minidisk, DVD (Digital Versatile Disc), etc.

[0106] It is also possible to realize some of the functions of data acquisition unit 11, lifestyle pattern identification unit 12, lifestyle pattern compilation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, data provision unit 16, and a control unit (not shown) with dedicated hardware and some with software or firmware. For example, the functions of data acquisition unit 11 and data provision unit 16 can be realized by processing circuit 101 as dedicated hardware, and the functions of lifestyle pattern identification unit 12, lifestyle pattern compilation unit 13, lifestyle rhythm estimation unit 14, change detection unit 15, and a control unit (not shown) can be realized by processor 104 reading and executing programs stored in memory 105. The storage unit 17 is configured by the memory 105 or a HDD or the like. The life rhythm estimation device 1 also includes devices such as a sensor 2 or an output device 3, an input interface device 102 and an output interface device 103 that perform wired or wireless communication.

[0107] As described above, according to the first embodiment, the lifestyle rhythm estimation device 1 is configured to include a data acquisition unit 11 that acquires daily activity-related data related to the subject's daily activity detected by the sensor 2; a lifestyle pattern identification unit 12 that compares the daily activity-related data acquired by the data acquisition unit 11 with a plurality of basic pattern data that each indicate different basic patterns of lifestyle patterns to identify the subject's lifestyle pattern during the lifestyle pattern identification period; a lifestyle pattern aggregation unit 13 that aggregates the subject's lifestyle pattern during the lifestyle rhythm estimation period based on the subject's lifestyle pattern identified by the lifestyle pattern identification unit 12 and lifestyle pattern aggregation conditions that set the aggregation unit and aggregation method for the subject's lifestyle pattern during the lifestyle rhythm estimation period; and a lifestyle rhythm estimation unit 14 that estimates the subject's lifestyle rhythm from the aggregation results of the subject's lifestyle pattern aggregated by the lifestyle pattern aggregation unit 13. Therefore, the lifestyle rhythm estimation device 1 can estimate a generalized lifestyle rhythm by estimating the lifestyle rhythm of the subject during the lifestyle rhythm estimation period, taking into account that people do not always follow a fixed lifestyle pattern during each of the multiple lifestyle pattern identification periods, but rather live with a variety of lifestyle patterns. In other words, the lifestyle rhythm estimation device 1 can generalize and grasp the lifestyle rhythm of the subject.

[0108] For example, the lifestyle pattern counting conditions may include a condition that the frequency of occurrence of each lifestyle pattern is counted, with each lifestyle pattern being used as a counting unit. In this case, in lifestyle rhythm estimation device 1, lifestyle pattern counting unit 13 counts the frequency of occurrence of each lifestyle pattern during a lifestyle rhythm estimation period based on the subject's lifestyle patterns identified by lifestyle pattern identification unit 12 and the lifestyle pattern counting conditions, and lifestyle rhythm estimation unit 14 estimates the subject's lifestyle rhythm from the frequency of occurrence of each lifestyle pattern. Therefore, the lifestyle rhythm estimation device 1 can estimate a generalized lifestyle rhythm by estimating the lifestyle rhythm of the subject during the lifestyle rhythm estimation period, taking into account that people do not always follow a fixed lifestyle pattern during each of the multiple lifestyle pattern identification periods, but rather live in a variety of lifestyle patterns. In other words, the lifestyle rhythm estimation device 1 can generalize and grasp the lifestyle rhythm of the subject. Furthermore, the lifestyle rhythm estimation device 1 can estimate the lifestyle rhythm specific to each individual.

[0109] For example, the lifestyle pattern counting conditions may specify that the frequency or proportion of occurrence of lifestyle patterns for each day of the week should be counted. In this case, in lifestyle rhythm estimation device 1, lifestyle pattern counting unit 13 counts the frequency or proportion of occurrence of lifestyle patterns for each day of the week during a lifestyle rhythm estimation period based on the subject's lifestyle patterns identified by lifestyle pattern identification unit 12 and the lifestyle pattern counting conditions, and lifestyle rhythm estimation unit 14 estimates the subject's lifestyle rhythm from the frequency or proportion of occurrence of lifestyle patterns for each day of the week. Therefore, the lifestyle rhythm estimation device 1 can estimate a generalized lifestyle rhythm by estimating the lifestyle rhythm of the subject during the lifestyle rhythm estimation period, taking into account that people do not always follow a fixed lifestyle pattern during each of the multiple lifestyle pattern identification periods, but rather live in a variety of lifestyle patterns. In other words, the lifestyle rhythm estimation device 1 can generalize and grasp the lifestyle rhythm of the subject. Furthermore, the lifestyle rhythm estimation device 1 can estimate the lifestyle rhythm specific to each individual.

[0110] For example, the lifestyle rhythm estimation period is a period that includes one or more cycles, each of which is made up of multiple days, and the lifestyle pattern aggregation conditions specify that the frequency or proportion of occurrence of daily lifestyle patterns is to be aggregated using each day in the cycle as the aggregation unit. In this case, in lifestyle rhythm estimation device 1, lifestyle rhythm estimation unit 14 estimates the subject's lifestyle rhythm from the frequency or proportion of occurrence of daily lifestyle patterns. Therefore, the lifestyle rhythm estimation device 1 can estimate a generalized lifestyle rhythm by estimating the lifestyle rhythm of the subject during the lifestyle rhythm estimation period, taking into account that people do not always follow a fixed lifestyle pattern during each of the multiple lifestyle pattern identification periods, but rather live in a variety of lifestyle patterns. In other words, the lifestyle rhythm estimation device 1 can generalize and grasp the lifestyle rhythm of the subject. Furthermore, the lifestyle rhythm estimation device 1 can estimate the lifestyle rhythm specific to each individual.

[0111] Furthermore, the lifestyle rhythm estimation device 1 can be configured to include a change detection section 15 that detects a change in the lifestyle rhythm of the subject estimated by the lifestyle rhythm estimation section 14. This enables the lifestyle rhythm estimation device 1 to detect data useful for managing the subject's physical and mental health. The lifestyle rhythm estimation device 1 also enables detection of disruptions to the subject's lifestyle. For example, when monitoring an elderly person living alone, the lifestyle rhythm estimation device 1 can detect abnormalities or estimate changes in the subject's lifestyle associated with dementia.

[0112] Furthermore, the lifestyle rhythm estimation device 1 can be configured to include a data providing unit 16 that outputs provided data relating to the lifestyle rhythm of the subject estimated by the lifestyle rhythm estimation unit 14. This allows the lifestyle rhythm estimation device 1 to provide data useful for improving the subject's physical and mental health management. Furthermore, the lifestyle rhythm estimation device 1 can present the estimated lifestyle rhythm specific to each individual.

[0113] For example, in life rhythm estimation device 1, data providing section 16 can be configured to output data indicating a change in the life rhythm of the subject detected by change detection section 15 as provided data. This enables the lifestyle rhythm estimation device 1 to provide the subject with data useful for improving the subject's physical and mental health. The lifestyle rhythm estimation device 1 can also present estimated or detected lifestyle disorders of the subject. For example, in the case of monitoring an elderly person living alone, the lifestyle rhythm estimation device 1 can present the elderly person's family with estimated abnormalities or changes in the subject's lifestyle associated with dementia.

[0114] Furthermore, for example, in the lifestyle rhythm estimation device 1, the data providing section 16 can be configured to output data explaining the details of the subject's lifestyle rhythm as provided data. This allows the lifestyle rhythm estimation device 1 to provide data that allows the subject to understand the subject's physical and mental health state. Furthermore, the lifestyle rhythm estimation device 1 can provide data that allows the content explaining the estimated individual-specific lifestyle rhythm to be understood.

[0115] Furthermore, for example, in the lifestyle rhythm estimation device 1, the data providing unit 16 can be configured to output data proposing measures to improve the lifestyle rhythm of the subject as provided data. This enables the life rhythm estimation device 1 to provide the subject with data that is useful for improving the subject's physical and mental health.

[0116] Furthermore, for example, in the lifestyle rhythm estimation device 1, the data providing unit 16 can be configured to compare the lifestyle rhythm of the subject with the lifestyle rhythms of others to extract characteristics of the lifestyle rhythm of the subject, and output data explaining the extracted characteristics as provided data. As a result, the lifestyle rhythm estimation device 1 can provide the subject with data that allows the subject to understand the subject's physical and mental health status, and can also provide the subject with data that is useful for improving the subject's physical and mental health.

[0117] Furthermore, for example, in the lifestyle rhythm estimation device 1, the data providing unit 16 can be configured to compare the lifestyle rhythm of the subject with a recommended lifestyle rhythm and output an index indicating the subject's health level as provided data. As a result, the lifestyle rhythm estimation device 1 can provide the subject with data that allows the subject to understand the subject's physical and mental health status, and can also provide the subject with data that is useful for improving the subject's physical and mental health.

[0118] Any of the components of the embodiments may be modified or omitted.

[0119] Various aspects of the present disclosure are summarized below as appendices.

[0120] (Appendix 1) a data acquisition unit that acquires daily behavior related data regarding the subject's daily behavior detected by the sensor; a lifestyle pattern identification unit that compares the daily activity related data acquired by the data acquisition unit with a plurality of basic pattern data each indicating a different basic pattern of a lifestyle, and identifies the subject's lifestyle pattern during a period for identifying the subject's lifestyle pattern; a lifestyle pattern counting unit that counts the lifestyle patterns of the subject during the life rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and lifestyle pattern counting conditions that set a counting unit and a counting method for the lifestyle patterns of the subject during the life rhythm estimation period; a life rhythm estimation unit that estimates a life rhythm of the subject from the result of the aggregation of the life patterns of the subject aggregated by the life pattern aggregation unit; A life rhythm estimation device comprising: (Appendix 2) The period for identifying a lifestyle pattern includes one day, a set time period, or a period consisting of multiple days. 2. A life rhythm estimation device according to claim 1. (Appendix 3) The lifestyle pattern tallying condition includes a condition that the occurrence frequency of each lifestyle pattern is tallied with the lifestyle pattern as the tallying unit, and the lifestyle pattern counting unit counts an appearance frequency of each of the lifestyle patterns during the lifestyle rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and the lifestyle pattern counting condition; The life rhythm estimation unit estimates the life rhythm of the subject based on the frequency of occurrence of each of the life patterns. 3. A life rhythm estimation device according to claim 1 or 2. (Appendix 4) The lifestyle pattern tallying condition is set to tally the occurrence frequency or ratio of the lifestyle pattern for each day of the week, with each day of the week being the tallying unit; the lifestyle pattern counting unit counts an appearance frequency or a ratio of the lifestyle pattern for each day of the week during the lifestyle rhythm estimation period based on the lifestyle pattern of the subject identified by the lifestyle pattern identifying unit and the lifestyle pattern counting condition; The life rhythm estimation unit estimates the life rhythm of the subject from an appearance frequency or a ratio of the life pattern for each day of the week. 3. A life rhythm estimation device according to claim 1 or 2. (Appendix 5) The life rhythm estimation period is a period including one or more cycles, each cycle consisting of a plurality of days, the lifestyle pattern tallying condition is set to tally the occurrence frequency or ratio of the lifestyle pattern for each day, with the day in the cycle being the tallying unit; The life rhythm estimation unit estimates the life rhythm of the subject from the occurrence frequency or ratio of the life pattern for each day. 3. A life rhythm estimation device according to claim 1 or 2. (Appendix 6) a change detection unit that detects a change in the life rhythm of the subject estimated by the life rhythm estimation unit; 6. A life rhythm estimation device according to any one of appendices 1 to 5, (Appendix 7) The change detection unit compares a first life rhythm of the subject estimated by the life rhythm estimation unit with a second life rhythm of the subject, and detects a change in the life rhythm of the subject. 7. A life rhythm estimation device according to claim 6. (Appendix 8) The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject by comparing a component ratio of the aggregation result of the life pattern indicated by the first life rhythm with a component ratio of the aggregation result of the life pattern indicated by the second life rhythm. 8. A life rhythm estimation device according to claim 7. (Appendix 9) The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject based on a ratio or a variance of each life pattern in a summary result of the plurality of life patterns, which is indicated by the life rhythm of the subject estimated for the plurality of life rhythm estimation periods of the time series estimated by the life rhythm estimation unit. 7. A life rhythm estimation device according to claim 6. (Appendix 10) The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject based on whether or not a new life pattern has appeared in the life rhythm of the subject. 7. A life rhythm estimation device according to claim 6. (Appendix 11) The life rhythm estimation period is a period including one or more cycles, each cycle consisting of a plurality of days, The change detection unit detects a change in the life rhythm of the subject based on the daily activity related data acquired by the data acquisition unit, depending on whether or not periodicity of the daily activity related data has changed. 7. A life rhythm estimation device according to claim 6. (Appendix 12) a data providing unit that outputs provided data regarding the subject's life rhythm estimated by the life rhythm estimation unit, The data providing unit outputs, as the provided data, data indicating the change in the life rhythm of the subject detected by the change detecting unit. 12. The life rhythm estimation device according to claim 6, wherein: (Appendix 13) a data providing unit that outputs provided data regarding the life rhythm of the subject estimated by the life rhythm estimation unit; 6. A life rhythm estimation device according to any one of appendices 1 to 5, (Appendix 14) The data providing unit outputs data explaining the content of the life rhythm of the subject as the provided data. 14. A life rhythm estimation device according to claim 13. (Appendix 15) The data providing unit outputs data proposing an improvement measure for the lifestyle rhythm of the subject as the provided data. 14. A life rhythm estimation device according to claim 13. (Appendix 16) The data providing unit compares the life rhythm of the subject with the life rhythm of another person to extract characteristics of the life rhythm of the subject, and outputs data explaining the extracted characteristics as the provided data. 14. A life rhythm estimation device according to claim 13. (Appendix 17) The data providing unit compares the life rhythm of the subject with a recommended life rhythm, and outputs an index indicating the health level of the subject as the provided data. 14. A life rhythm estimation device according to claim 13. (Appendix 18) Computer, a data acquisition unit that acquires daily behavior related data regarding the subject's daily behavior detected by the sensor; a lifestyle pattern identification unit that compares the daily activity related data acquired by the data acquisition unit with a plurality of basic pattern data each indicating a different basic pattern of a lifestyle, and identifies the subject's lifestyle pattern during a period for identifying the subject's lifestyle pattern; a lifestyle pattern counting unit that counts the lifestyle patterns of the subject during the life rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and lifestyle pattern counting conditions that set a counting unit and a counting method for the lifestyle patterns of the subject during the life rhythm estimation period; a life rhythm estimation unit that estimates a life rhythm of the subject from the result of the aggregation of the life patterns of the subject aggregated by the life pattern aggregation unit; A lifestyle rhythm estimation program to function as a [Explanation of symbols]

[0121] 1 Life rhythm estimation device, 11 Data acquisition unit, 12 Life pattern identification unit, 13 Life pattern aggregation unit, 14 Life rhythm estimation unit, 15 Change detection unit, 16 Data provision unit, 17 Memory unit, 2 Sensor, 3 Output device, 101 Processing circuit, 102 Input interface device, 103 Output interface device, 104 Processor, 105 Memory.

Claims

1. a data acquisition unit that acquires daily behavior related data regarding the subject's daily behavior detected by the sensor; a lifestyle pattern identification unit that compares the daily activity related data acquired by the data acquisition unit with a plurality of basic pattern data each indicating a different basic pattern of a lifestyle, and identifies the subject's lifestyle pattern during a period for identifying the subject's lifestyle pattern; a lifestyle pattern counting unit that counts the lifestyle patterns of the subject during the life rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and lifestyle pattern counting conditions that set a counting unit and a counting method for the lifestyle patterns of the subject during the life rhythm estimation period; a life rhythm estimation unit that estimates a life rhythm of the subject from the result of the aggregation of the life patterns of the subject aggregated by the life pattern aggregation unit; A life rhythm estimation device comprising:

2. The period for identifying a lifestyle pattern includes one day, a set time period, or a period consisting of multiple days.

2. The life rhythm estimation device according to claim 1.

3. The lifestyle pattern tallying condition includes a condition that the occurrence frequency of each lifestyle pattern is tallied with the lifestyle pattern as the tallying unit, and the lifestyle pattern counting unit counts an appearance frequency of each of the lifestyle patterns during the lifestyle rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and the lifestyle pattern counting condition; The life rhythm estimation unit estimates the life rhythm of the subject based on the frequency of occurrence of each of the life patterns.

3. The life rhythm estimation device according to claim 1 or 2.

4. The lifestyle pattern tallying condition is set to tally the occurrence frequency or ratio of the lifestyle pattern for each day of the week, with each day of the week being the tallying unit; the lifestyle pattern counting unit counts an appearance frequency or a ratio of the lifestyle pattern for each day of the week during the lifestyle rhythm estimation period based on the lifestyle pattern of the subject identified by the lifestyle pattern identifying unit and the lifestyle pattern counting condition; The life rhythm estimation unit estimates the life rhythm of the subject from an appearance frequency or a ratio of the life pattern for each day of the week.

3. The life rhythm estimation device according to claim 1 or 2.

5. the life rhythm estimation period is a period including one or more cycles, each cycle consisting of a plurality of days; the lifestyle pattern tallying condition is set to tally the occurrence frequency or ratio of the lifestyle pattern for each day, with the day in the cycle being the tallying unit; The life rhythm estimation unit estimates the life rhythm of the subject from the occurrence frequency or ratio of the life pattern for each day.

3. The life rhythm estimation device according to claim 1 or 2.

6. a change detection unit that detects a change in the life rhythm of the subject estimated by the life rhythm estimation unit; 2. The life rhythm estimation device according to claim 1.

7. The change detection unit compares a first life rhythm of the subject estimated by the life rhythm estimation unit with a second life rhythm of the subject, and detects a change in the life rhythm of the subject.

7. The life rhythm estimation device according to claim 6.

8. The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject by comparing a component ratio of the aggregation result of the life pattern indicated by the first life rhythm with a component ratio of the aggregation result of the life pattern indicated by the second life rhythm.

8. The life rhythm estimation device according to claim 7.

9. The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject based on a ratio or a variance of each life pattern in a summary result of the plurality of life patterns, which is indicated by the life rhythm of the subject estimated for the plurality of life rhythm estimation periods of the time series estimated by the life rhythm estimation unit.

7. The life rhythm estimation device according to claim 6.

10. The lifestyle rhythm is represented by a summary result of a plurality of the lifestyle patterns, The change detection unit detects a change in the life rhythm of the subject based on whether or not a new life pattern has appeared in the life rhythm of the subject.

7. The life rhythm estimation device according to claim 6.

11. the life rhythm estimation period is a period including one or more cycles, each cycle consisting of a plurality of days; The change detection unit detects a change in the life rhythm of the subject based on the daily activity related data acquired by the data acquisition unit, depending on whether or not periodicity of the daily activity related data has changed.

7. The life rhythm estimation device according to claim 6.

12. a data providing unit that outputs provided data regarding the subject's life rhythm estimated by the life rhythm estimation unit, The data providing unit outputs, as the provided data, data indicating the change in the life rhythm of the subject detected by the change detecting unit.

7. The life rhythm estimation device according to claim 6.

13. a data providing unit that outputs provided data regarding the life rhythm of the subject estimated by the life rhythm estimation unit; 2. The life rhythm estimation device according to claim 1.

14. The data providing unit outputs data explaining the content of the life rhythm of the subject as the provided data.

14. The life rhythm estimation device according to claim 13.

15. The data providing unit outputs data proposing an improvement measure for the lifestyle rhythm of the subject as the provided data.

14. The life rhythm estimation device according to claim 13.

16. The data providing unit compares the life rhythm of the subject with the life rhythm of another person to extract characteristics of the life rhythm of the subject, and outputs data explaining the extracted characteristics as the provided data.

14. The life rhythm estimation device according to claim 13.

17. The data providing unit compares the life rhythm of the subject with a recommended life rhythm, and outputs an index indicating the health level of the subject as the provided data.

14. The life rhythm estimation device according to claim 13.

18. Computer, a data acquisition unit that acquires daily behavior related data regarding the subject's daily behavior detected by the sensor; a lifestyle pattern identification unit that compares the daily activity related data acquired by the data acquisition unit with a plurality of basic pattern data each indicating a different basic pattern of a lifestyle, and identifies the subject's lifestyle pattern during a period for identifying the subject's lifestyle pattern; a lifestyle pattern counting unit that counts the lifestyle patterns of the subject during the life rhythm estimation period based on the lifestyle patterns of the subject identified by the lifestyle pattern identifying unit and lifestyle pattern counting conditions that set a counting unit and a counting method for the lifestyle patterns of the subject during the life rhythm estimation period; a life rhythm estimation unit that estimates a life rhythm of the subject from the result of the aggregation of the life patterns of the subject aggregated by the life pattern aggregation unit; A lifestyle rhythm estimation program to function as a

Images