Estimation device, estimation method, and program

Abstract

An estimation device and the like are provided that can estimate state transitions taking past states into consideration even when there is no chronological data on the same object. [Solution] An estimation device according to one aspect of the present disclosure includes a receiving means for receiving input of a data set for each stage, a first estimation means for estimating a first simultaneous distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage, a second estimation means for estimating a second simultaneous distribution based on a transition from the estimated first simultaneous distribution to a distribution of data in a third stage that is a stage subsequent to the second stage, and a calculation means for calculating a state transition probability regarding the transition of data from the second stage to the third stage based on the second simultaneous distribution.

Description

[Technical Field]

[0001] The present disclosure relates to an estimation device, an estimation method, and a program. [Background technology]

[0002] BACKGROUND OF THE INVENTION Techniques exist for future analysis of information about people in fields such as healthcare.

[0003] Patent document 1 discloses a technology in which the probability that a subject will remain in the same sleep state and the probability that a subject will transition from one sleep state to another are experimentally determined in advance, and the determined probabilities are used to estimate the subject's sleep depth. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-089267 Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, information about a target person is predicted by estimating a transition destination state from the target person's current state based on the probability of state transition. Here, for example, when predicting health-related data, it may be necessary to consider not only the current state but also past states. In other words, it may be necessary to consider past states when estimating state transitions.

[0006] Furthermore, in Patent Document 1, when estimating a state transition related to a person, the probability of the transition is calculated based on data accumulated in advance. For example, time-series data obtained by observing the same person for a predetermined period is used to calculate such a probability. However, it may be difficult to accumulate such time-series data related to the same object. If there is no time-series data related to the same object, it is difficult to experimentally determine the probability of the transition.

[0007] One of the objectives of the present disclosure has been made in consideration of the above-mentioned problems, and is to provide an estimation device, etc. that is capable of estimating state transitions taking past states into account even when there is no longitudinal data on the same object. [Means for solving the problem]

[0008] An estimation device according to one aspect of the present disclosure includes a receiving means for receiving input of a data set for each stage, a first estimation means for estimating a first simultaneous distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage, a second estimation means for estimating a second simultaneous distribution based on a transition from the estimated first simultaneous distribution to a distribution of data in a third stage that is a stage subsequent to the second stage, and a calculation means for calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second simultaneous distribution.

[0009] An estimation method according to one aspect of the present disclosure receives input of a data set for each stage, estimates a first joint distribution based on a transition from the distribution of data in a first stage to the distribution of data in a second stage that is a stage subsequent to the first stage, estimates a second joint distribution based on a transition from the estimated first joint distribution to the distribution of data in a third stage that is a stage subsequent to the second stage, and calculates a state transition probability for the transition of data from the second stage to the third stage based on the second joint distribution.

[0010] A program according to one aspect of the present disclosure causes a computer to perform the following processes: accepting input of a data set for each stage; estimating a first joint distribution based on a transition from the distribution of data in a first stage to the distribution of data in a second stage that is a stage subsequent to the first stage; estimating a second joint distribution based on a transition from the estimated first joint distribution to the distribution of data in a third stage that is a stage subsequent to the second stage; and calculating a state transition probability for the transition of data from the second stage to the third stage based on the second joint distribution. [Effects of the Invention]

[0011] According to the present disclosure, even when there is no time-series data on the same person, it is possible to estimate a state transition that takes past states into consideration. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a first block diagram illustrating an example of a functional configuration of an estimation device according to the present disclosure. [Figure 2] 1 is a first flowchart illustrating an example of the operation of the estimation device of the present disclosure. [Figure 3] FIG. 2 is a second block diagram illustrating an example of the functional configuration of the estimation device of the present disclosure. [Figure 4] FIG. 10 is a diagram illustrating an example of a probability density distribution according to the present disclosure. [Figure 5] FIG. 1 is a diagram illustrating an image of solving a transition from one probability density distribution to another probability density distribution as an optimal transportation problem according to the present disclosure. [Figure 6] 10 is a second flowchart illustrating an example of the operation of the estimation device of the present disclosure. [Figure 7] FIG. 10 is a third block diagram illustrating an example of the functional configuration of the estimation device of the present disclosure. [Figure 8] 10 is a third flowchart illustrating an example of the operation of the estimation device of the present disclosure. [Figure 9]FIG. 4 is a fourth block diagram illustrating an example of the functional configuration of the estimation device of the present disclosure. [Figure 10] 10 is a fourth flowchart illustrating an example of the operation of the estimation device of the present disclosure. [Figure 11] 10 is a fifth flowchart illustrating an example of the operation of the estimation device of the present disclosure. [Figure 12] FIG. 2 is a block diagram illustrating an example of a hardware configuration of a computer device that realizes the estimation device of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0014] First Embodiment An outline of the estimation device of the first embodiment will be described.

[0015] The estimation device of the present disclosure uses accumulated data to calculate state transition probabilities between data. In the present disclosure, an example of target data is health data, which is data related to a person's health. The health data may be, for example, values ​​of test items in a health checkup or information related to a person's exercise habits. However, the health data is not limited to these examples.

[0016] The health data may be accumulated in advance, for example. In this case, the accumulated health data does not have to be longitudinal data resulting from continuous observation of a specific individual over a predetermined period of time. The accumulated health data may be data for each of multiple individuals at a predetermined point in time. For example, the results of health checkups for 10,000 individuals in a year t may be accumulated as health data. Note that, although the present disclosure mainly describes an example in which the estimation device estimates data transitions based on health data, the target data is not limited to this example.

[0017] 1 is a first block diagram illustrating an example of the functional configuration of the estimation device 100. As illustrated in FIG. 1, the estimation device 100 includes a receiving unit 110, a first estimation unit 120, a second estimation unit 130, and a calculation unit 140.

[0018] The receiving unit 110 receives input of a dataset. The dataset may be health data of multiple people. For example, the dataset may be the results of health checkups conducted on multiple people over a one-year period. In this way, the dataset may be data related to the results of health checkups at a specific time point for each of multiple people. The dataset may be stored in advance in a storage device (not shown). The storage device may be a device included in the estimation device 100, or may be an external device communicatively connected to the estimation device 100.

[0019] Datasets are classified into stages. A stage may be information indicating a stratum when a dataset is stratified. In other words, a stage can also be information indicating a condition when population data is classified into subsets based on a predetermined condition. For example, if a dataset is health data, a dataset for each stage may be health data for each age group. More specifically, if the health data is data indicating the blood glucose levels of multiple people, a dataset for each stage may include data indicating the blood glucose levels of people in their teens, data indicating the blood glucose levels of people in their twenties, ..., and data indicating the blood glucose levels of people in their eighties. In other words, in this example, the dataset includes data indicating the blood glucose levels for each age group of ten years. In this way, the stages may be ordered. For example, the stage following the stage for people in their teens is the stage for people in their twenties. Note that the age groups may be any age group.

[0020] In this way, the receiving unit 110 receives input of a data set for each stage. The receiving unit 110 is an example of a receiving means.

[0021] The first estimation unit 120 estimates a joint distribution based on a transition of data between predetermined stages. Specifically, the first estimation unit 120 estimates a first joint distribution based on a transition from the distribution of data in a first stage to the distribution of data in a second stage.

[0022] Assume that the health data is data indicating the blood glucose levels of multiple individuals. Assume also that the first stage is the age group in their 40s, and the second stage is the age group in their 50s. That is, the second stage is a stage subsequent to the first stage. In this case, the first estimation unit 120 estimates, for example, a transition from the blood glucose level distribution of individuals in their 40s to the blood glucose level distribution of individuals in their 50s. The first estimation unit 120 may perform the estimation using an algorithm for the optimal transportation problem. That is, the first estimation unit 120 estimates a plausible transition from a probability distribution indicating the probability that individuals in their 40s with each blood glucose level exist to a probability distribution indicating the probability that individuals in their 50s with each blood glucose level exist. In this estimation, a distribution indicating the correspondence between the possible values ​​of the random variables in each probability distribution and their probabilities is estimated. The distribution indicating this correspondence is called a joint distribution. The joint distribution estimated by the first estimation unit 120 is called a first joint distribution.

[0023] In this way, the first estimating unit 120 estimates a first joint distribution based on a transition from the distribution of data in the first stage to the distribution of data in the second stage, which is a stage subsequent to the first stage. The first estimating unit 120 is an example of a first estimating means.

[0024] The second estimation unit 130 estimates a joint distribution based on a transition from the estimated first joint distribution to the distribution of data in a third stage. The third stage is a stage subsequent to the second stage. For example, if the second stage is the age group in their 50s and the third stage is the age group in their 60s, the third stage is a stage subsequent to the second stage.

[0025] Assume that the first joint distribution is a joint distribution based on the transition from the blood glucose level distribution of people in their 40s to the blood glucose level distribution of people in their 50s, as described above. Assume that the third stage is the age group of people in their 60s. In this case, the second estimation unit 130, for example, estimates the transition from the first joint distribution to the blood glucose level distribution of people in their 60s. In this case, the second estimation unit 130 may perform the estimation using an algorithm for the optimal transportation problem, similar to the first estimation unit 120.

[0026] That is, the second estimation unit 130 estimates a joint distribution that indicates the correspondence between the possible values ​​of random variables and their probabilities in the first joint distribution and the probability distribution that indicates the probability that there are people in their 60s with each blood glucose level. The joint distribution estimated by the second estimation unit 130 is referred to as the second joint distribution. It can be said that the second joint distribution indicates the transition from the blood glucose level distribution of people in their 50s to the blood glucose level distribution of people in their 60s, taking into account the blood glucose level distribution of people in their 40s.

[0027] In this way, the second estimating unit 130 estimates the second joint distribution based on the transition from the estimated first joint distribution to the distribution of data in the third stage, which is the stage after the second stage. The second estimating unit 130 is an example of a second estimating means.

[0028] Then, the calculation unit 140 calculates the state transition probability regarding the transition of data from the second stage to the third stage based on the second joint distribution. The calculation unit 140 is an example of a calculation means.

[0029] Next, an example of the operation of the estimation device 100 will be described with reference to Fig. 2. In this disclosure, each step in a flowchart will be represented by a number assigned to the step, such as "S1".

[0030] FIG. 2 is a flowchart illustrating an example of the operation of the estimation device 100.

[0031] The receiving unit 110 receives input of a data set for each stage (S1).

[0032] The first estimation unit 120 estimates a first joint distribution based on a transition from the distribution of data in the first stage to the distribution of data in a second stage that is a stage subsequent to the first stage (S2).

[0033] The second estimation unit 130 estimates a second joint distribution based on a transition from the estimated first joint distribution to the distribution of data in a third stage, which is a stage subsequent to the second stage (S3).

[0034] The calculation section 140 calculates the state transition probability regarding the transition of data from the second stage to the third stage based on the second joint distribution (S4).

[0035] As described above, the estimation device 100 of the first embodiment receives input of a data set for each stage. The estimation device 100 estimates a first joint distribution based on a transition from the distribution of data in the first stage to the distribution of data in a second stage, which is a stage subsequent to the first stage. The estimation device 100 then estimates a second joint distribution based on a transition from the estimated first joint distribution to the distribution of data in a third stage, which is a stage subsequent to the second stage. The estimation device 100 then calculates a state transition probability for the transition of data from the second stage to the third stage, based on the second joint distribution.

[0036] The estimation device 100 estimates a transition from a data distribution in a first stage to a data distribution in a second stage using a first joint distribution based on the transition from the data distribution in a first stage to the data distribution in a second stage. This allows the estimation device 100 to take the data in the first stage into consideration when estimating the transition from the data distribution in the second stage to the data distribution in the third stage. Furthermore, the estimation device 100 estimates the first joint distribution. In other words, the estimation device 100 does not use a method that requires data over time on the same person, such as experimentally determining the probability of a transition. In other words, the estimation device 100 can estimate a state transition that takes past states into consideration, even when there is no data over time on the same person.

[0037] <Second embodiment> Next, an estimation device according to a second embodiment will be described. In the second embodiment, a further example of the estimation device described in the first embodiment will be described. In the second embodiment, an example in which the estimation device 100 estimates data transitions based on health data will also be mainly described, but the target data is not limited to this example. Note that some of the content overlapping with the first embodiment will be omitted.

[0038] [Details of the estimation device 100] 3 is a block diagram showing an example of the functional configuration of the estimation device 100. The estimation device 100 includes a receiving unit 110, a first estimation unit 120, a second estimation unit 130, and a calculation unit 140. The estimation device 100 may also include an acquisition unit 150 and a classification unit 160. The estimation device 100 may also include a storage device 190. The storage device 190 may be a device included in the estimation device 100, or may be an external device communicatively connected to the estimation device 100.

[0039] The estimation device 100 is, for example, a device provided in a terminal device such as a personal computer. The terminal device is a device operated by a user. The estimation device 100 is not limited to this example, and may be a device realized in a server device communicably connected to the terminal device via a wired or wireless network. The estimation device 100 may perform various processes in response to instructions from the terminal device.

[0040] The estimation device 100 may also be communicatively connected to other devices via a wired or wireless network. For example, the estimation device 100 may be capable of communicating with an external server device that stores health data. The external server device may be, for example, a device managed by a hospital, a local government, or a company.

[0041] The receiving unit 110 receives input of a dataset for each stage. At this time, the dataset is stored in the storage device 190. For example, the receiving unit 110 may receive, as input of a dataset, reading a dataset stored in the storage device 190 in accordance with an instruction from a terminal device.

[0042] Health data is stored in storage device 190. Receiving section 110 may receive the health data as a data set. The health data is acquired by acquiring section 150.

[0043] The acquiring unit 150 acquires health data. Specifically, the acquiring unit 150 acquires the health data from an external server device that manages the health data. For example, assume that the external server device manages the health checkup results of 10,000 people in year t. The acquiring unit 150 acquires the health checkup results of the 10,000 people in year t from the external server device as health data. The health data may be information corresponding to the test items in the health checkup. Furthermore, the health data is the results of health checkups taken by each of the 10,000 people at a certain point in time in year t. In this way, the data set may not be data showing changes over time in the same subject, but may be data measured at a single point in time for each of multiple subjects. The acquiring unit 150 stores the acquired health data in the storage device 190.

[0044] The method of acquiring health data is not limited to this example. For example, a recording medium storing health data may exist. In this case, the terminal device reads the health data from the recording medium. Then, acquisition unit 150 may acquire the health data read by the terminal device.

[0045] In this way, acquiring section 150 acquires health data, which is data relating to the health of each of a plurality of people at a predetermined time point. Acquiring section 150 is an example of an acquiring means.

[0046] The health data is processed by, for example, classifying unit 160. Then, the processed health data may be stored in storage device 190. Classifying unit 160 processes the data acquired by acquiring unit 150 into a data set according to conditions. For example, classifying unit 160 classifies the health data by age group. For example, classifying unit 160 classifies the health data by age group of 10 years. However, this is not a limitation, and classifying unit 160 may classify the health data by any age interval. For example, classifying unit 160 may classify the health data by one year of age.

[0047] In this case, the classification unit 160 may extract specific data from the health data and classify the extracted data by age group. For example, assume that the health data includes information indicating height, weight, blood pressure, blood glucose level, HbA1c, and BMI (Body Math Index). In this case, the classification unit 160 may classify the data indicating blood glucose level and BMI from the health data by age group.

[0048] The classification conditions and the data to be extracted may be information according to instructions from a terminal device. That is, a user operating the terminal device inputs information indicating the classification conditions and the data to be extracted into the terminal device. The terminal device transmits the input information to the estimation device 100. The classification unit 160 processes the health data using the information indicating the classification conditions and the data to be extracted transmitted from the terminal device.

[0049] The classification unit 160 also generates a distribution related to the acquired data. Specifically, the classification unit 160 generates a probability density distribution for each condition based on the acquired data. For example, the classification unit 160 generates a distribution in which data indicating blood glucose levels and BMI are plotted for each age group. The distribution generated at this time is a two-dimensional distribution related to blood glucose levels and BMI. The classification unit 160 then generates a probability density distribution indicating the probability of existence of each value of blood glucose levels and BMI. The one-dimensional data value is expressed as x i Let the data values ​​of other dimensions be x j Then the probability density distribution is p([x i ,x j ]) In this way, the classifier 160 generates a probability density distribution for each age group regarding the acquired health data.

[0050] In this case, the classification unit 160 may classify the data in each distribution into data groups. In this case, the probability density distribution is a distribution indicating the existence probability of each data group. FIG. 4 is a diagram showing an example of a probability density distribution. The probability density distribution shown in FIG. 4 has blood glucose level and BMI as its axes. In addition, the example of FIG. 4 shows 64 cells. These cells are data groups into which data indicating each person's blood glucose level and BMI is classified. The existence probability of each data group is shown. In this way, the classification unit 160 may classify the data in each health data distribution for each age group into data groups. Note that the classification unit 160 may generate a two-dimensional or higher-dimensional distribution. For example, the classification unit 160 may generate a probability density distribution with blood glucose level, BMI, and average number of steps per day as its axes. The probability density distribution divided into cells as shown in FIG. 4 corresponds to a marginal distribution based on each value of the health data. The probability density distribution discussed below may be the probability distribution shown in FIG. 4 or may not take the form of a marginal distribution.

[0051] The receiving unit 110 receives the health data categorized by condition as a data set for each stage. For example, the receiving unit 110 may receive a probability density distribution for the health data for each age group as described above as a data set for each stage.

[0052] The first estimation unit 120 estimates a first joint distribution between stages. Specifically, the first estimation unit 120 estimates the first joint distribution based on a transition from the distribution of health data in a first age group to the distribution of health data in a second age group. Here, the second age group is an age group after the first age group. Hereinafter, the distribution of health data in the first age group will be referred to as the first distribution. Also, the distribution of health data in the second age group will be referred to as the second distribution.

[0053] The first estimation unit 120 estimates the transition from the first distribution to the second distribution using an algorithm for the optimal transportation problem (hereinafter referred to as the optimal transportation algorithm). The optimal transportation algorithm is an algorithm that finds a transportation method that optimizes the cost required to transition from a given probability distribution to another probability distribution.

[0054] Specifically, for distributions μ and v in the probability space X, the Cartesian product X 2 The distribution π in is a coupling when the following equations 1 and 2 hold.

[0055]

number

[0056]

number

[0057] Let the total number of couplings be Π(μ,v). Also, let c(x,y) be the cost function for transporting element x included in distribution μ to element y included in distribution v. In this case, for example, in the following equation (3), the coupling that minimizes the cost is called optimal transport. Note that the distribution π in this case corresponds to the joint distribution.

[0058]

number

[0059] In other words, the optimal transportation algorithm can calculate a set of data before transportation and data at the destination that optimizes the cost of transportation from the data distribution in the first stage to the data distribution in the second stage.

[0060] The first estimation unit 120 estimates a first joint distribution based on a transition from a first distribution, which is the distribution of health data in a first age group, to a second distribution, which is the distribution of health data in a second age group that is an age group subsequent to the first age group. In this case, the first estimation unit 120 solves the transition from the first distribution to the second distribution as an optimal transportation problem. Here, the first distribution and the second distribution are probability density distributions. Specifically, the first estimation unit 120 solves the transition from the probability density distribution for health data in the first age group to the probability density distribution for health data in the second age group as an optimal transportation problem.

[0061] FIG. 5 is a diagram illustrating an example of solving the transition from one probability density distribution to another as an optimal transportation problem. FIG. 5 shows a probability density distribution for health data in a first age group and a probability density distribution for health data in a second age group. Solving the transition from the first distribution to the second distribution as an optimal transportation problem by the first estimation unit 120 corresponds to estimating which of the cells in the probability density distribution for the second age group has a higher probability of transitioning from each cell in the probability density distribution for the first age group. In other words, the first estimation unit 120 estimates a first joint distribution based on the transition from each data group in the first distribution to each data group in the second distribution.

[0062] For example, let μ be the probability density distribution for health data in a first age group, and v be the probability density distribution for health data in a second age group. In this case, the first estimation unit 120 estimates the first joint distribution π∈Π(μ,v) using Equation 3.

[0063] The second estimation unit 130 estimates a second joint distribution based on a transition from the estimated first joint distribution to the distribution of health data in a third age group, which is an age group subsequent to the second age group. Hereinafter, the distribution of health data in the third age group will also be referred to as the third distribution.

[0064] The second estimating unit 130, like the first estimating unit 120, estimates the transition from the first joint distribution to the distribution of data in the third stage using an optimal transportation algorithm. That is, the second estimating unit 130 estimates the second joint distribution using an optimal transportation algorithm that calculates pairs of data before transportation and data at the transportation destination, which optimizes the cost of transporting data from the first joint distribution to the distribution of data in the third stage.

[0065] For example, the second estimation unit 130 solves the transition from the first joint distribution to the third distribution as an optimal transportation problem. The third distribution is a probability density distribution. At this time, a new cost function is defined. Let ξ be the third distribution (i.e., the probability density distribution related to health data in the third age group). Let z be an element of the distribution ξ. At this time, c((x, y), z) is defined as a new cost function. The direct product X 2 Let Π(π,ξ) be the total coupling of distributions ρ in × X. Then, the second estimating unit 130 estimates the second joint distribution ρ∈Π(π,ξ) based on Equation 3 in which the new cost function is defined.

[0066] The calculation unit 140 estimates state transition probabilities related to transitions in health data based on the joint distribution. In the optimal transportation problem, finding the optimal coupling (joint distribution) is equivalent to finding the optimal state transition. In other words, it is possible to find the state transition probability from the joint distribution. For example, the calculation unit 140 calculates the state transition probability ρ(z|x,y) from the second age group to the third age group based on the second joint distribution ρ.

[0067] As described above, the state transition probability ρ(z|x,y) is information based on the result of solving the optimal transportation problem for the transition from the first joint distribution to the third distribution. Therefore, the state transition probability ρ(z|x,y) also takes into account the health data for the first age group. In other words, the calculation of the state transition probability ρ(z|x,y) does not assume Markovianity for the state transition. That is, the calculation unit 140 calculates the non-Markovian state transition probability ρ(z|x,y) from the second age group to the third age group, taking into account the health data for the first age group. Note that in this example, an example of solving the optimal transportation problem has been described using a probability density distribution (marginal distribution) in which the first and second distributions are divided into cells as shown in FIG. 4. Examples of solving the optimal transportation problem are not limited to this example. For example, the first and second distributions may be distributions in which the health data is not classified into data groups, and each value of the health data and its existence probability are indicated.

[0068] In this way, the calculation section 140 can calculate the state transition probability from the distribution of data in the second stage to the distribution of data in the third stage based on the second joint distribution.

[0069] Similarly, the calculation unit 140 can calculate the state transition probability π(y|x) from the first age group to the second age group based on the first joint distribution π. That is, the calculation unit 140 can calculate the state transition probability from the distribution of data in the first stage to the distribution of data in the second stage based on the first joint distribution. The calculation unit 140 may store the calculated state transition probability in the storage device 190.

[0070] [Example of operation of the estimation device 100] Next, an example of the operation of the estimating device 100 will be described with reference to FIG.

[0071] FIG. 6 is a second flowchart illustrating an example of the operation of estimation device 100. Specifically, FIG. 6 is a flowchart illustrating an example of the operation of estimation device 100 when calculating a state transition probability between predetermined stages. In this operation example, it is assumed that data indicating blood glucose levels and BMI are acquired as health data. This operation example also illustrates an example in which estimation device 100 calculates a state transition probability from a health data distribution in the 50s age group to a health data distribution in the 60s age group.

[0072] Acquiring unit 150 acquires health data (S101). For example, acquiring unit 150 acquires health data from an external server device. Acquiring unit 150 then stores the health data in storage device 190. Classifying unit 160 processes the health data (S102). For example, classifying unit 160 classifies the health data by age group. Then, classifying unit 160 generates a probability density distribution for each age group in the 10s for the health data. At this time, the probability density distribution is a distribution that indicates the probability of existence of each value of blood glucose level and BMI.

[0073] Receiving unit 110 receives input of health data for each age group (S103). For example, receiving unit 110 receives reading of health data stored in storage device 190 as input of health data.

[0074] First estimation unit 120 estimates a first joint distribution based on a transition from the distribution of health data in a first age group (first distribution) to the distribution of health data in a second age group (second distribution) (S104). Here, the first age group is the 40s. The second age group is the 50s. That is, first estimation unit 120 estimates a first joint distribution based on a transition from the probability density distribution for health data in the 40s to the probability density distribution for health data in the 50s.

[0075] The second estimation unit 130 estimates a second joint distribution based on a transition from the first joint distribution to a distribution of health data in a third age group (third distribution) (S105). Here, the third age group is the 60s. That is, the second estimation unit 130 estimates the second joint distribution based on a transition from the joint distribution that takes into account the transition of health data from the 40s to the 50s to the probability density distribution for health data in the 60s.

[0076] Then, calculation unit 140 estimates the state transition probability based on the estimated second joint distribution (S106). Specifically, calculation unit 140 calculates the state transition probability from the probability density distribution related to the health data of people in their 50s to the probability density distribution related to the health data of people in their 60s based on the second joint distribution.

[0077] Note that this operational example is merely an example. That is, the operation of the estimation device 100 of the present disclosure is not limited to this example. Furthermore, the estimation device 100 does not necessarily have to estimate state transition probabilities for consecutive stages. In this operational example, an example is shown in which the state transition probability is calculated from health data for people in their 50s to health data for people in their 60s, taking into account health data for people in their 40s, based on health data categorized by age group in their teens. This example is not limiting, and for example, the estimation device 100 may calculate the state transition probability from health data for people in their 50s to health data for people in their 70s, taking into account health data for people in their 40s. Furthermore, the estimation device 100 may calculate the state transition probability from health data for people in their 50s to health data for people in their 60s, taking into account health data for people in their 30s.

[0078] As described above, the estimation device 100 of the second embodiment receives input of a data set for each stage. The estimation device 100 also estimates a first joint distribution based on a transition from the distribution of data in the first stage to the distribution of data in the second stage, which is a stage subsequent to the first stage. The estimation device 100 also estimates a second joint distribution based on a transition from the estimated first joint distribution to the distribution of data in the third stage, which is a stage subsequent to the second stage. The estimation device 100 then calculates a state transition probability for the transition of data from the second stage to the third stage, based on the second joint distribution.

[0079] Specifically, for example, estimation device 100 receives health data for each age group, which is data related to the health of multiple people at a given time point, as a data set for each stage. Estimation device 100 then estimates a first joint distribution based on a transition from a first distribution, which is the distribution of health data in a first age group, to a second distribution, which is the distribution of health data in a second age group that is an age group subsequent to the first age group. Estimation device 100 then estimates a second joint distribution based on a transition from the estimated first joint distribution to a third distribution, which is the distribution of health data in a third age group that is an age group subsequent to the second age group. Estimation device 100 then calculates a state transition probability for the transition of health data from the second age group to the third age group based on the second joint distribution.

[0080] In this way, estimation device 100 estimates the transition from the health data distribution in the first age group to the health data distribution in the second age group using a first joint distribution based on the transition from the health data distribution in the first age group to the health data distribution in the second age group. This allows estimation device 100 to take the health data in the first age group into consideration when estimating the transition from the health data distribution in the second age group to the health data distribution in the third age group. Furthermore, estimation device 100 estimates the first joint distribution. In other words, estimation device 100 does not use a method that requires longitudinal data on the same person, such as experimentally determining the probability of transition. In other words, estimation device 100 can estimate state transitions that take past states into account, even when longitudinal data on the same person is unavailable.

[0081] Furthermore, by calculating the state transition probability in this manner, it is possible to predict, for example, based on the health data of a specified person whose age corresponds to the second age group, what value the health data will have when that specified person reaches the age of the third age group.

[0082] The estimating apparatus 100 may estimate the first joint distribution using an optimal transportation algorithm that calculates pairs of data before transportation and data at a destination, optimizing the cost of transporting data from the distribution of data in the first stage to the distribution of data in the second stage. The estimating apparatus 100 may estimate the second joint distribution using an optimal transportation algorithm that calculates pairs of data before transportation and data at a destination, optimizing the cost of transporting data from the first joint distribution to the distribution of data in the third stage.

[0083] This allows the estimation device 100 to calculate state transition probabilities from optimal transport between distributions even if the data targets at each stage are not the same. That is, the estimation device 100 can estimate state transitions even when there is no time-series data for the same target.

[0084] <Third embodiment> Next, an estimation device according to a third embodiment will be described. In the third embodiment, an example will be described in which the health state of a target person is predicted based on the calculated state transition probability. In the third embodiment, an example in which the estimation device estimates data transitions based on health data will also be mainly described, but the target data is not limited to this example. Note that some of the content overlapping with the first and second embodiments will not be described.

[0085] [Details of Estimation Device 101] The estimation device 101 is a device obtained by adding additional functional units to the estimation device 100. FIG. 7 is a block diagram showing an example of the functional configuration of the estimation device 101. The estimation device 101 includes a reception unit 110, a first estimation unit 120, a second estimation unit 130, a calculation unit 140, an acquisition unit 150, and a classification unit 160. The estimation device 101 may also include a prediction unit 170. The estimation device 101 may also include a storage device 190.

[0086] Like the estimation device 100, the estimation device 101 may be a device provided in a terminal device, or may be a device implemented in a server device communicatively connected to the terminal device via a wired or wireless network.

[0087] Estimation device 101 predicts future health data for a target person using pre-calculated state transition probabilities. In this embodiment, the first age group is the 40s, the second age group is the 50s, and the third age group is the 60s. The state transition probability from health data in the 50s to health data in the 60s is then calculated.

[0088] The acquisition unit 150 acquires health data of the target person. For example, assume that the state transition probability is calculated from the probability density distribution related to the blood glucose level and BMI. At this time, the acquisition unit 150 acquires health data indicating the blood glucose level and BMI of the target person.

[0089] Prediction unit 170 is an example of a prediction means. Prediction unit 170 predicts the transition of the target person's health data. In other words, prediction unit 170 predicts the future value of the target person's health data based on the target person's health data. Specifically, prediction unit 170 predicts the value of the target person's health data when the target person reaches an age corresponding to a third age group. In this case, the second age group is the age group corresponding to the target person's age.

[0090] For example, suppose the target person is 51 years old. In this case, the target person's age corresponds to the second age group. The prediction unit 170 identifies which data group the target person's health data falls into among the probability density distributions for health data in the 50s. The prediction unit 170 then predicts which data group the identified data group will transition to among the probability density distributions for health data in the 60s, based on the state transition probability.

[0091] In this way, the prediction unit 170 predicts the data group in the third distribution, which is the transition destination based on the state transition probability of the data group in the second distribution corresponding to the health data of the target person, as the health data when the target person reaches the age of the third age group.

[0092] [Example of operation of the estimation device 101] Next, an example of the operation of the estimation device 101 will be described with reference to FIG.

[0093] Fig. 8 is a third flowchart illustrating an example of the operation of the estimation device 101. Specifically, Fig. 8 is a flowchart illustrating an example of the operation of the estimation device 101 when predicting the health state of a target person.

[0094] The acquisition unit 150 acquires health data of the target person (S201). For example, the acquisition unit 150 acquires the health data of the target person from a terminal device.

[0095] The prediction unit 170 identifies a data group in the second distribution corresponding to the health data of the target person (S202). For example, the prediction unit 170 identifies which data group the health data of the target person falls into in the probability density distribution for health data in the second age group.

[0096] Then, prediction unit 170 predicts the health data of the target person when he / she reaches the third age group based on the state transition probability (S203). Specifically, prediction unit 170 predicts the health data of the target person when he / she reaches the third age group based on the state transition probability of the identified data group in the probability density distribution for the health data in the third age group.

[0097] Note that this operation example is merely an example, and the operation of the estimation device 101 of the present disclosure is not limited to this example.

[0098] As described above, the estimation device 101 of the third embodiment acquires health data of a target person. In this case, the second age group corresponds to the target person's age. The estimation device 101 then predicts the data group in the third distribution, which is the transition destination of the data group in the second distribution corresponding to the target person's health data based on the state transition probability, as the health data of the target person when they reach the age of the third age group. This allows the estimation device 101 to predict the target person's future health state.

[0099] <Fourth embodiment> Next, an estimation device according to a fourth embodiment will be described. In the fourth embodiment, a further example of predicting the health state of a target person based on calculated state transition probabilities will be described. In the fourth embodiment, an example in which the estimation device estimates data transitions based on health data will also be mainly described, but the target data is not limited to this example. Note that some of the content overlapping with the first, second, and third embodiments will not be described.

[0100] [Details of the estimation device 102] The estimation device 102 is a device obtained by adding additional functional units to the estimation device 101. FIG. 9 is a block diagram showing an example of the functional configuration of the estimation device 102. The estimation device 102 includes a reception unit 110, a first estimation unit 120, a second estimation unit 130, a calculation unit 140, an acquisition unit 150, a classification unit 160, and a prediction unit 170. The estimation device 101 may also include a generation unit 180. The estimation device 102 may also include a storage device 190.

[0101] Like the estimation device 101, the estimation device 102 may be a device provided in a terminal device, or may be a device implemented in a server device communicatively connected to the terminal device via a wired or wireless network.

[0102] The estimation device 102 calculates state transition probabilities between stages in advance. Then, the estimation device 102 generates a learning model using the calculated state transition probabilities. The estimation device 102 also uses the generated learning model to predict future health data for the target person. In this embodiment, the stage of generating the learning model is referred to as the generation phase, and the stage of making predictions is referred to as the prediction phase.

[0103] (Generation phase) It is assumed that a data set for each stage is stored in advance in storage device 190. For example, it is assumed that a probability density distribution regarding health data for each age group is stored in storage device 190.

[0104] The receiving unit 110 receives input of a probability density distribution related to health data for each age group. In this embodiment, the probability density distribution is a distribution related to health data for each age group of 10. Furthermore, it is assumed that there are eight types of probability density distributions, from the teens to the 80s.

[0105] The first estimation unit 120 estimates a first joint distribution regarding adjacent stages. Specifically, the first estimation unit 120 estimates a first joint distribution based on the transition of the data distribution from the first stage to the stage adjacent to the first stage among the data sets for each stage. For example, the first estimation unit 120 estimates a first joint distribution based on the transition from the probability density distribution regarding the health data of the 10-year-old group to the probability density distribution regarding the health data of the 20-year-old group. Note that the first estimation unit 120 may estimate a first joint distribution based on the transition from the probability density distribution regarding the health data of the 20-year-old group to the probability density distribution regarding the health data of the 30-year-old group. Similarly, the first estimation unit 120 may estimate a first joint distribution based on the transition of the data distribution for each set of adjacent stages, based on the probability density distribution regarding the health data between adjacent stages. Hereinafter, the first joint distribution (N < M) based on the transition of the probability density distribution regarding the health data in the age group from the Nth generation to the Mth generation is referred to as the first joint distribution between the Nth and Mth generations.

[0106] The second estimation unit 130 estimates a second joint distribution based on the transition from the first joint distribution to the distribution of health data in other age groups. For example, the second estimation unit 130 estimates a second joint distribution based on the transition from the first joint distribution between the 10-year-old and 20-year-old groups to the distribution of health data in the 30-year-old age group.

[0107] Furthermore, the second estimation unit 130 uses the estimated second joint distribution to estimate a further joint distribution. Specifically, the estimated second joint distribution is regarded as the first joint distribution. Then, the second estimation unit 130 estimates a second joint distribution based on a transition from the joint distribution regarded as the first joint distribution to the distribution of data in an adjacent stage. For example, suppose the second estimation unit 130 estimates a second joint distribution based on a transition from the first joint distribution between the teens and twenties to the distribution of health data in the thirties age group. The estimated second joint distribution is regarded as the first joint distribution between the teens and thirties. Therefore, the second estimation unit 130 estimates a second joint distribution based on a transition from the first joint distribution between the teens and thirties to the distribution of health data in the forties age group. Similarly, since the second joint distribution is regarded as the first joint distribution between the teens and forties, the second estimation unit 130 estimates a second joint distribution based on a transition from the first joint distribution between the teens and forties to a distribution of health data in the fifties age group. The second estimation unit 130 continues this process until it estimates a second joint distribution based on a transition to a distribution of health data in the eighties age group. In other words, the second estimation unit 130 continues this process until it estimates a second joint distribution based on a transition to a distribution of data in the final stage of the dataset.

[0108] In this way, the first estimating unit 120 first estimates a first joint distribution based on a transition in data distribution between adjacent stages. The second estimating unit 130 estimates a second joint distribution based on a transition from the first joint distribution to the data distribution in the stage adjacent to the first joint distribution. The second estimating unit 130 then regards the estimated second joint distribution as the first joint distribution and repeats the process of estimating a second joint distribution based on a transition to the data distribution in the adjacent stage until the adjacent stage becomes the last stage.

[0109] The calculation unit 140 calculates the state transition probability regarding the transition of data between each stage, based on each of the first joint distribution and the second distribution.

[0110] The generation unit 180 generates a machine learning model. Specifically, the generation unit 180 generates a prediction model that outputs data in another stage, which is a transition destination of data in one stage, based on the calculated state transition probability. The prediction model corresponds to a machine learning model that learns the relationship between the distribution of data in one stage and the distribution of data in another stage. Alternatively, the generation unit 180 may generate a prediction model that outputs a data group of data in another stage, which is a transition destination of a data group of data distribution in one stage, based on the calculated state transition probability. The generated prediction model is a machine learning model that receives age and health data as input, and outputs health data to which the input health data will transition after a predetermined period, or a group of such data.

[0111] In this way, the generation unit 180 generates a machine learning model that learns the relationship between the data in the stage before the transition and the data in the stage after the transition based on the state transition probability. The generation unit 180 is an example of a generation means.

[0112] (Prediction phase) The acquisition unit 150 acquires health data of the target person.

[0113] The prediction unit 170 uses a machine learning model to estimate the future health state of the target person. Specifically, the prediction unit 170 inputs the target person's health data and the target person's age into the machine learning model. The machine learning model outputs health data for age groups above the target person's age. In other words, the machine learning model outputs health data for the target person when they reach the age they will be at after a predetermined period of time has passed. The prediction unit 170 predicts the health data as the target person's future health data.

[0114] For example, suppose a target person is 51 years old. Suppose health data for the target person when they reach their 70s is to be predicted. In this case, prediction unit 170 inputs, for example, information indicating that the target person is 51 years old and the target person's health data into the machine learning model. At this time, the machine learning model outputs health data for the 70s age group, which is the transition destination of the input health data. Prediction unit 170 outputs the output health data as health data for the target person when they reach their 70s.

[0115] The machine learning model may be a model that outputs health data after a specific period of time has passed for input health data. For example, the machine learning model may output health data for a target person when they reach an age group 20 years from now. The machine learning model may also be a model that outputs trends in health data up to a specific period of time has passed. For example, the model may output trends in health data for a target person until they reach an age group 20 years from now.

[0116] [Example of operation of the estimation device 102] Next, an example of the operation of the estimation device 102 will be described with reference to FIGS.

[0117] Fig. 10 is a fourth flowchart illustrating an example of the operation of estimation device 102. Specifically, Fig. 10 is a flowchart illustrating an example of the operation of estimation device 102 in the generation phase. In the operation example of Fig. 10, it is assumed that a probability density distribution related to health data for each age group is stored in advance in storage device 190.

[0118] The receiving unit 110 receives an input of a probability density distribution related to health data for each age group (S301). For example, the receiving unit 110 receives an input of a probability density distribution related to health data for each age group.

[0119] The first estimation unit 120 estimates a first joint distribution between adjacent stages (S302). For example, the first estimation unit 120 estimates a first joint distribution based on the transition of the distribution of data between adjacent stages, based on the probability density distribution of health data between the teens age group, which is the first stage, and the twenties age group, which is the adjacent stage.

[0120] The second estimation unit 130 estimates a second joint distribution for each stage after the adjacent stage (S303). Specifically, the second estimation unit 130 estimates a second joint distribution based on a transition from the first joint distribution to the distribution of data in the stage adjacent to the adjacent stage. Then, the second estimation unit 130 regards the estimated second joint distribution as the first joint distribution, and repeats the process of estimating a second joint distribution based on a transition to the distribution of data in the adjacent stage until the adjacent stage becomes the last stage.

[0121] The calculation unit 140 calculates state transition probabilities related to data transitions between stages (S304). Then, the generation unit 180 generates a machine learning model based on the calculated state transition probabilities (S305). Specifically, the generation unit 180 generates a machine learning model that learns the relationship between data in a stage before the transition and data in a stage after the transition based on the state transition probabilities.

[0122] Fig. 11 is a fifth flowchart illustrating an example of the operation of the estimating device 102. Specifically, Fig. 11 is a flowchart illustrating an example of the operation of the estimating device 102 in the prediction phase.

[0123] The acquisition unit 150 acquires the health data of the target person (S401). For example, the acquisition unit 150 acquires the health data of the target person from a terminal device.

[0124] Prediction unit 170 uses the machine learning model to estimate the future health state of the target person (S402). Specifically, prediction unit 170 inputs information indicating the target person's age and health data into the machine learning model. The health data is then output by the machine learning model. Prediction unit 170 outputs the output health data as the health data of the target person when they reach the age after a predetermined period of time has passed.

[0125] Note that this operation example is merely an example, and the operation of the estimation device 102 of the present disclosure is not limited to this example.

[0126] As described above, the estimation device 102 of the fourth embodiment estimates a first joint distribution based on a transition of data distributions between adjacent stages. The estimation device 102 also estimates a second joint distribution based on a transition from the first joint distribution to a data distribution in a stage adjacent to the first joint distribution. The estimation device 102 then regards the estimated second joint distribution as the first joint distribution and repeats the process of estimating a second joint distribution based on a transition to a data distribution in an adjacent stage, until the adjacent stage becomes the last stage. The estimation device 102 then calculates state transition probabilities for data transitions between each stage based on each second joint distribution. The estimation device 102 then generates a machine learning model that learns the relationship between data in a stage before the transition and data in a stage after the transition based on the state transition probabilities.

[0127] <Modification> In the present disclosure, examples have been described in which the estimation device estimates data transitions based on health data. That is, examples have been described in which the estimation device is used in the fields of healthcare or medicine. However, examples in which the estimation device is applied are not limited to these. For example, the estimation device may also be applied to estimating state transitions of various machines.

[0128] For example, when measurement data measured on the operating state of a machine is acquired, the estimation device may accept the measurement data for each state based on aging changes, from a state in which the machine is operating normally to a state in which the machine is broken, as a data set for each stage. The estimation device may also estimate a first joint distribution based on the distribution of the measurement data between each state, and estimate a second joint distribution based on the transition from the first joint distribution to the distribution of data in another state. The estimation device may then calculate a state transition probability for the transition between each state.

[0129] <Example of hardware configuration of estimation device> The hardware constituting the estimation devices of the first, second, third, and fourth embodiments described above will be described. FIG. 12 is a block diagram showing an example of the hardware configuration of a computer device constituting the estimation device in each embodiment. The estimation device and estimation method described in each embodiment and each modified example are realized in a computer device 90. For example, the estimation device etc. described in each embodiment and each modified example may have the hardware configuration shown in FIG. 12.

[0130] 12, a computer device 90 includes a processor 91, a RAM (Random Access Memory) 92, a ROM (Read Only Memory) 93, a storage device 94, an input / output interface 95, a bus 96, and a drive device 97. Note that the estimation device and the like may be realized by a plurality of electric circuits.

[0131] The storage device 94 stores a program (computer program) 98. The processor 91 executes the program 98 of the present estimation device using the RAM 92. Specifically, for example, the program 98 includes a program that causes a computer to execute processes shown in FIGS. 2, 6, 8, 10, and 11, etc. The processor 91 executes the program 98 to realize the functions of each component of the present estimation device. The program 98 may be stored in the ROM 93. Alternatively, the program 98 may be recorded on the recording medium 80 and read out using the drive device 97, or may be transmitted to the computer device 90 from an external device (not shown) via a network (not shown).

[0132] The input / output interface 95 exchanges data with peripheral devices (such as a keyboard, a mouse, and a display device) 99. The input / output interface 95 functions as a means for acquiring or outputting data. The bus 96 connects each component.

[0133] There are various variations in the method for realizing the estimation device. For example, each component included in the estimation device can be realized as a dedicated device. Furthermore, the estimation device can be realized based on a combination of multiple devices.

[0134] The scope of each embodiment also includes a processing method for recording a program for realizing each configuration of the function of each embodiment on a recording medium, reading the program recorded on the recording medium as code, and executing it on a computer. In other words, a computer-readable recording medium is also included in the scope of each embodiment. Furthermore, a recording medium on which the above-mentioned program is recorded and the program itself are also included in each embodiment.

[0135] The recording medium may be, but is not limited to, a floppy disk, a hard disk, an optical disk, a magneto-optical disk, a CD (Compact Disc)-ROM, a magnetic tape, a non-volatile memory card, or a ROM. The programs recorded on the recording medium are not limited to standalone programs that execute processes, but also include programs that run on an OS (Operating System) in cooperation with other software and functions of an expansion board.

[0136] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0137] Furthermore, the above-described embodiments and modifications can be combined as appropriate.

[0138] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0139] <Additional Notes> [Appendix 1] A receiving means for receiving input of a data set for each stage; a first estimation means for estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; a second estimation means for estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; a calculation means for calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution, Estimation device.

[0140] [Appendix 2] the receiving means receives health data, which is data on the health of each of a plurality of people at a predetermined time point, for each age group, as a data set for each stage; the first estimation means estimates the first joint distribution based on a transition from a first distribution, which is a distribution of the health data in a first age group, to a second distribution, which is a distribution of the health data in a second age group that is an age group subsequent to the first age group; the second estimation means estimates the second joint distribution based on a transition from the estimated first joint distribution to a third distribution, which is a distribution of the health data in a third age group that is an age group subsequent to the second age group; the calculation means calculates a state transition probability regarding a transition of the health data from the second age group to the third age group based on the second joint distribution. 10. The estimation apparatus of claim 1.

[0141] [Appendix 3] an acquisition means for acquiring the health data, which is data relating to the health of each of a plurality of persons at a predetermined time; a classification means for classifying data in each of the distributions of the health data for each age group into data groups; the first estimation means estimates the first joint distribution based on transitions from each data group in the first distribution to each data group in the second distribution; the second estimation means estimates the second joint distribution based on a transition from each data group in the first joint distribution to each data group in the third distribution; 10. The estimation device of claim 2.

[0142] [Appendix 4] a prediction means for predicting a transition of the health data of a target person based on the state transition probability; the acquisition means acquires health data of the target person; the second age group is an age group corresponding to the age of the target person, the prediction means predicts a data group in the third distribution, which is a transition destination of the data group in the second distribution corresponding to the health data of the target person based on the state transition probability, as health data of the target person when the target person reaches the age of the third age group; 10. The estimation device of claim 3.

[0143] [Appendix 5] The classification means generates a probability density distribution for each age group regarding the acquired health data; each of the first distribution, the second distribution, and the third distribution is a probability density distribution; The probability density distribution indicates the existence probability of each of the data groups. 10. The estimation device of claim 3.

[0144] [Appendix 6] the first estimation means estimates the first joint distribution using an optimal transportation algorithm that calculates a set of data before transportation and data at a transportation destination, which optimizes the cost of transportation from the distribution of data in the first stage to the distribution of data in the second stage; the second estimation means estimates the second joint distribution using an optimal transportation algorithm that calculates a set of data before transportation and data at a destination, optimizing the cost of transportation from the first joint distribution to the distribution of data in the third stage; 10. The estimation apparatus of claim 1.

[0145] [Appendix 7] The dataset is data regarding the results of health checkups of each of a plurality of people at a predetermined time point. 10. The estimation apparatus of claim 1.

[0146] [Appendix 8] Further comprising a generating means for generating a machine learning model; the first estimation means estimates the first joint distribution based on a transition of data distribution between adjacent stages; the second estimation means estimates the second joint distribution based on a transition from the first joint distribution to a distribution of data in an adjacent stage subsequent to the adjacent stage; the second estimation means regards the estimated second joint distribution as a first joint distribution, and repeats a process of estimating the second joint distribution based on a transition to a distribution of data in an adjacent stage until the adjacent stage becomes the last stage; the calculation means calculates state transition probabilities regarding data transitions between stages based on each of the second joint distributions; the generation means generates a machine learning model that learns a relationship between data in a stage before the transition and data in a stage after the transition based on state transition probabilities. 10. The estimation apparatus of claim 1.

[0147] [Appendix 9] Accepts input of a dataset for each stage, Estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution; Estimation method.

[0148] [Appendix 10] A process for accepting input of a dataset for each stage; A process of estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; a process of estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution; program.

[0149] Furthermore, some or all of the configurations described in Supplementary Notes 2 to 8, which are dependent on Supplementary Note 1, may also be dependent on Supplementary Notes 9 and 10 in the same dependent relationship as Supplementary Notes 2 to 8. Furthermore, within the scope of each of the above-mentioned embodiments, some or all of the configurations described as Supplements may be made dependent on various hardware, software, various recording means for recording software, or systems. [Explanation of symbols]

[0150] 100, 101, 102 Estimator 110 Reception 120 1st estimation part 130 Second estimation part 140 Calculation Unit 150 Acquisition Department 160 Classification Department 170 Prediction Department 180 Generation part 190 Storage device

Claims

1. A receiving means for receiving input of a data set for each stage; a first estimation means for estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; a second estimation means for estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; a calculation means for calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution, Estimation device.

2. the receiving means receives health data, which is data on the health of each of a plurality of people at a predetermined time point, for each age group, as a data set for each stage; the first estimation means estimates the first joint distribution based on a transition from a first distribution, which is a distribution of the health data in a first age group, to a second distribution, which is a distribution of the health data in a second age group that is an age group subsequent to the first age group; the second estimation means estimates the second joint distribution based on a transition from the estimated first joint distribution to a third distribution, which is a distribution of the health data in a third age group that is an age group subsequent to the second age group; the calculation means calculates a state transition probability regarding a transition of the health data from the second age group to the third age group based on the second joint distribution. The estimation device according to claim 1 .

3. an acquisition means for acquiring the health data, which is data relating to the health of each of a plurality of persons at a predetermined time; a classification means for classifying data in each of the distributions of the health data for each age group into data groups; the first estimation means estimates the first joint distribution based on a transition from each data group in the first distribution to each data group in the second distribution; the second estimation means estimates the second joint distribution based on a transition from each data group in the first joint distribution to each data group in the third distribution; The estimation device according to claim 2 .

4. a prediction means for predicting a transition of the health data of a target person based on the state transition probability; the acquisition means acquires health data of the target person; the second age group is an age group corresponding to the age of the target person, the prediction means predicts a data group in the third distribution, which is a transition destination of the data group in the second distribution corresponding to the health data of the target person based on the state transition probability, as health data of the target person when the target person reaches an age in the third age group. The estimation device according to claim 3 .

5. The classification means generates a probability density distribution for each age group regarding the acquired health data; each of the first distribution, the second distribution, and the third distribution is a probability density distribution; The probability density distribution indicates the existence probability of each of the data groups. The estimation device according to claim 3 .

6. the first estimation means estimates the first joint distribution using an optimal transportation algorithm that calculates a set of data before transportation and data at a transportation destination, which optimizes the cost of transportation from the distribution of data in the first stage to the distribution of data in the second stage; the second estimation means estimates the second joint distribution using an optimal transportation algorithm that calculates a set of data before transportation and data at a transportation destination, optimizing the cost of transportation from the first joint distribution to the distribution of data in the third stage; The estimation device according to claim 1 .

7. The dataset is data regarding the results of health checkups of each of a plurality of people at a predetermined time point. The estimation device according to claim 1 .

8. Further comprising a generating means for generating a machine learning model; the first estimation means estimates the first joint distribution based on a transition of data distribution between adjacent stages; the second estimation means estimates the second joint distribution based on a transition from the first joint distribution to a distribution of data in an adjacent stage subsequent to the adjacent stage; the second estimation means regards the estimated second joint distribution as a first joint distribution, and repeats a process of estimating the second joint distribution based on a transition to a distribution of data in an adjacent stage until the adjacent stage becomes the last stage; the calculation means calculates state transition probabilities regarding data transitions between stages based on each of the second joint distributions; the generation means generates a machine learning model that learns a relationship between data in a stage before the transition and data in a stage after the transition based on state transition probabilities. The estimation device according to claim 1 .

9. Accepts input of a dataset for each stage, Estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution; Estimation method.

10. A process for accepting input of a dataset for each stage; A process of estimating a first joint distribution based on a transition from a distribution of data in a first stage to a distribution of data in a second stage that is a stage subsequent to the first stage; a process of estimating a second joint distribution based on a transition from the estimated first joint distribution to a distribution of data in a third stage that is a stage subsequent to the second stage; calculating a state transition probability regarding a transition of data from the second stage to the third stage based on the second joint distribution; program.

Images