Information Processing Method

Abstract

An information processing device 100 of the present invention comprises: a clustering unit 121 that classifies combination information representing combinations of a plurality of kinds of measures that have been implemented for a subject person at each time, into any of a plurality of previously set clusters; a first model generation unit 122 that generates a first model on the basis of state information representing the state of the subject person and the cluster into which the combinations of the plurality of kinds of measures implemented at each time for the subject person have been classified, the first model outputting a cluster corresponding to the state information; and a second model generation unit 123 that generates a second model on the basis of the state information, the cluster, and the combination information at each time, the second model outputting combination information with respect to the state information and information based on the cluster.

Description

[Technical Field]

[0001] The present invention relates to an information processing method, an information processing device, and a program. [Background technology]

[0002] There is an increasing demand for personalized medicine, which determines the treatment selection according to the condition of each patient when treating patients. Personalized medicine aims to provide treatment tailored to each patient using personal genetic information, medical information, etc. Furthermore, treatments may be configured by combining multiple treatments, and it is necessary to select the optimal treatment according to the patient's condition. Here, Patent Document 1 describes the use of a computer to select a treatment. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2016-514291 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when there are a large number of treatment options, the number of combinations of multiple treatments becomes enormous, making it difficult to select an appropriate treatment option for a patient. Furthermore, the more combinations of treatment options there are, the fewer examples of each combination of treatment options there are, making it difficult to refer to past examples. This problem is not limited to treatment, but also arises when training, exercising, dieting, etc., making it difficult to select a combination of multiple menus according to a person's condition. In other words, it is difficult to select a combination of multiple measures that can be implemented according to the target person's condition.

[0005] Therefore, the object of the present invention is to provide an information processing method that can solve the above-mentioned problem of the difficulty in selecting a combination of multiple measures that can be implemented depending on the condition of the target person. [Means for solving the problem]

[0006] An information processing method according to one aspect of the present invention includes: Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; The structure is as follows.

[0007] Furthermore, an information processing device according to one aspect of the present invention includes: a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; Equipped with The structure is as follows.

[0008] Furthermore, a program according to one aspect of the present invention includes: In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Execute the process, The structure is as follows. [Effects of the Invention]

[0009] By configuring the present invention as described above, it is possible to easily select a combination of multiple measures that can be implemented depending on the state of the target person. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 3] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 4] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 5] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 6] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 7] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 8]FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 9] FIG. 2 is a diagram illustrating data processing by the information processing device disclosed in FIG. [Figure 10] 2 is a flowchart showing the operation of the information processing device disclosed in FIG. [Figure 11] 2 is a flowchart showing the operation of the information processing device disclosed in FIG. [Figure 12] FIG. 10 is a diagram showing how data is processed by an information processing device according to a second embodiment of the present invention. [Figure 13] FIG. 10 is a diagram showing how data is processed by an information processing device according to a second embodiment of the present invention. [Figure 14] FIG. 10 is a block diagram showing the hardware configuration of an information processing device according to a third embodiment of the present invention. [Figure 15] FIG. 10 is a block diagram showing the configuration of an information processing device according to a third embodiment of the present invention. [Figure 16] 10 is a flowchart showing the operation of an information processing device according to a third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] <Embodiment 1> A first embodiment of the present invention will be described with reference to Fig. 1 to Fig. 11. Fig. 1 is a diagram for explaining the configuration of an information processing device, and Fig. 2 to Fig. 11 are diagrams for explaining the processing operation of the information processing device.

[0012] [Configuration] The information processing device 10 of the present invention is used to select a treatment method consisting of a combination of multiple treatments depending on the condition of each patient when treating a patient. Accordingly, the information processing device 10 is also used to generate a model for selecting a treatment method consisting of a combination of multiple treatments using existing patient data in advance. That is, the information processing device 10 has a learning function that learns existing patient data to generate a model, and a selection function that performs processing to select a future treatment method for the patient using the generated model. However, the information processing device 10 of the present invention may also be used to select a combination of multiple measures (e.g., treatments, menus, actions, suggestions) that can be implemented depending on the condition of a target person, not limited to treatment, in training, exercise, diet, etc.

[0013] The information processing device 10 is configured with one or more information processing devices each having a calculation device and a storage device. As shown in Fig. 1, the information processing device 10 includes an input unit 11, a clustering unit 12, a first learning unit 13, a second learning unit 14, and an output unit 15, which are constructed by the calculation device executing a program. The information processing device 10 also includes a data storage unit 16 and a model storage unit 17, which are formed in the storage device. Each component will be described in detail below.

[0014] The input unit 11 requests patient data from the data management device 20, accepts the input of the patient data, and stores the patient data in the data storage unit 16. Specifically, when generating a model, the input unit 11 accepts and stores, as learning data, patient data including the time (e.g., month and date) of treatment of a patient (subject person) who has already received multiple treatments, a treatment history (combination information) representing a combination of multiple types of treatments (measures) received by the patient, and a condition history (condition information) representing the patient's condition at that time. An example of patient data will now be described with reference to FIG. 2. In this figure, it is assumed that there is a treatment history A and a condition history X at times t1, t2, . . . , tn during a period T. In this figure, it is assumed that the treatment history A represents, for example, a combination of three types of treatments, i.e., treatments a, b, and c, at time t1, and a combination of three types of treatments, i.e., treatments a, d, and g, at time t2. The condition history X represents, for example, the patient's body temperature, blood pressure, heart rate, etc. at each time. The condition history X may include information such as the patient's age, height, weight, and medical history.

[0015] Furthermore, when performing a process of selecting a future treatment method for a patient, the input unit 11 requests and accepts input of the patient's patient data as selection data. For example, the input unit 11 accepts and stores input of patient data including the current time (e.g., date) and a condition history (condition information) that indicates the patient's condition at that time. As described above, the condition history includes information such as the patient's body temperature, blood pressure, heart rate, age, height, weight, and medical history at each time.

[0016] The clustering unit 12 performs a process of classifying the treatment history A included in the patient data stored as learning data into one of a plurality of preset clusters. Specifically, the clustering unit 12 converts the treatment history A into a vector expression with the type of treatment as an element, and classifies the treatment history A into one of a plurality of preset clusters according to the characteristics of the vector expression. Here, a specific example of the process by the clustering unit 12 will be described with reference to FIGS. 3 and 4.

[0017] First, the process of converting a patient's treatment history A into a vector representation will be described with reference to Fig. 3. Fig. 3 shows the treatment history A of a certain patient over a period T. For example, if there are 26 selectable types of treatments a to z, the clustering unit 12 converts the history into a treatment history vector AV having 26 elements, where elements of treatments that have been performed have a value of 1 (gray circles in Fig. 3) and elements of treatments that have not been performed have a value of 0 (white circles in Fig. 3). As an example, if the treatment history A at time t1 is a combination of three types of treatments, such as treatments a, b, and c, the clustering unit 12 converts the history into a treatment history vector AV1 having elements of all selectable treatments (a to z), where elements of treatments that have been performed have a value of 1 and elements of treatments that have not been performed have a value of 0. Furthermore, if the treatment history A at time t2 is a combination of three types of treatments, i.e., a, b, and g, the clustering unit 12 converts it into a treatment history vector AV2, with all selectable treatments (a to z) as elements, such that the element values ​​for treatments a, b, and g that have been performed are 1, and the element values ​​for treatments that have not been performed are 0.

[0018] Next, referring to FIG. 4, a process of classifying the patient's treatment history vector AV into clusters C is performed. Here, clusters C have, for example, three elements, and three types are set, where one of the elements has a value of 1 (gray square mark in FIG. 4) and the other elements have a value of 0 (white square mark in FIG. 4). In this way, the number of types of clusters C is set to three, which is fewer than the 26 types of treatments described above. However, the number of treatments and the number of clusters are not limited to the above numbers. The clustering unit 12 classifies the treatment history vector AV into one of three clusters C according to the characteristics of the vector array of the treatment history vector AV, for example, according to the distribution of elements with a value of 1 in the treatment history vector AV. As an example, the clustering unit 12 classifies the treatment history vector AV1 into cluster C1, in which the first element on the left side has a value of 1, because elements with a value of 1 are concentrated in the left region. However, the clustering unit 12 may classify the treatment history vector AV into cluster C using any method.

[0019] As described above, the clustering unit 12 converts each of the treatment histories A of a patient at each time into a treatment history vector AV and classifies them into cluster C. Then, the clustering unit 12 converts the treatment histories A of a patient at each time included in the patient data of a plurality of patients acquired as learning data into a treatment history vector AV and classifies them into cluster C.

[0020] The first learning unit 13 (first model generation unit) performs machine learning using the patient's condition history X for each time period and the cluster C into which the treatment history X given to the patient is classified, and generates a clustering model M1 (first model) that outputs a new cluster from a new condition history of the patient. Here, with reference to FIG. 5 , the process of generating the clustering model M1 will be described. The first learning unit 13 inputs the patient's condition history X, the patient's treatment history vector AV, and the cluster C at each time point t into the first learning unit 13, thereby learning the relationship between these and generating a clustering model M1 that outputs a new cluster in response to the input of a new condition history. The first learning unit 13 then stores the generated clustering model M1 in the model storage unit 17.

[0021] The second learning unit 14 (second model generation unit) performs machine learning using the patient's condition history X at each time, a treatment history vector AV of the treatment history X administered to the patient, and a cluster C into which the treatment history X is classified, to generate a restoration model M2 (second model) that outputs a new treatment proposal (combination information). Here, with reference to FIG. 6, the process of generating the restoration model M2 will be described. Specifically, the second learning unit 14 inputs the patient's condition history X, the patient's treatment history vector AV, the cluster C, and a cluster centroid vector CV (cluster feature information) that represents the features of the cluster C as a vector having multiple types of treatment as elements into the second learning unit 14, thereby learning the relationship between these. Then, as a result of the learning, the second learning unit 14 generates a restoration model M2 that outputs a new treatment proposal in response to an input of the patient's new condition history and a cluster centroid vector corresponding to a cluster output by inputting the new condition history into the clustering model M1.

[0022] Here, the above-mentioned cluster centroid vector is a vector whose elements are all treatments, and the value of the element corresponding to the treatment that represents the characteristics of cluster C is set to 1. For example, as shown in FIG. 7, a cluster centroid vector CV corresponding to each cluster C is set in advance. As an example, the cluster centroid vector has 26 elements, which is the number of treatment types. For cluster C1, in which the first element on the left has a value of 1, a cluster centroid vector CV1 is associated with the cluster C1, in which elements with a value of 1 are concentrated in the left-hand region. Note that the association between cluster C and cluster centroid vector CV as shown in FIG. 7 can be achieved by associating the cluster centroid vector CV with the identification information of cluster C. Therefore, in the above-mentioned second learning unit 14, during machine learning as shown in FIG. 6, instead of inputting the vector information of cluster C, it is sufficient to input the identification information of cluster C and the corresponding cluster centroid vector. That is, the second learning unit 14 generates a restoration model M2 that outputs a new treatment proposal by inputting the patient's condition history X, the patient's treatment history vector AV, the identification information of cluster C, and the cluster centroid vector CV corresponding to the identification information of cluster C at each time t.

[0023] The output unit 15 is executed when performing a process for selecting a future treatment method for a patient. Specifically, the output unit 15 first acquires a condition history representing the patient's current condition, which is included in the patient data of the patient accepted as selection data. As described above, the condition history includes information such as the patient's body temperature, blood pressure, heart rate, age, height, weight, and medical history. Then, as shown in FIG. 8 , the output unit 15 reads out the clustering model M1 generated as described above and stored in the model storage unit 17, and inputs the patient's condition history into the clustering model M1. Then, a new cluster C corresponding to the input condition history is output from the clustering model M1, and the output unit 15 acquires the new cluster C.

[0024] Next, the output unit 15 reads out the restoration model M2 generated as described above and stored in the model storage unit 17, and inputs the patient's condition history and information based on the new cluster C output from the clustering model M1 to the restoration model M2. Specifically, as shown in Fig. 9, the output unit 15 inputs the cluster centroid vector associated with the identification information of the new cluster C and the patient's condition history to the restoration model M2. Then, the restoration model M2 outputs a treatment proposal B that represents, in vector form, a combination of new treatments corresponding to the input condition history and cluster centroid vector, and the output unit 15 outputs this.

[0025] The output unit 15 extracts a plurality of treatments to be performed from the treatment proposal B expressed as a vector obtained as described above, and outputs them to a display screen, etc. This allows a person to whom the treatment proposal is displayed to refer to the treatment proposal and set a treatment method.

[0026] [Operation] Next, the operation of the information processing device 10 described above will be described with reference to the flowcharts of FIGS. 10 and 11. First, the operation when the information processing device 10 generates a model will be described with reference to FIG. 10. The information processing device 10 requests past patient data from the data management device 20 and acquires the patient data as learning data (step S1). As shown in FIG. 2, the patient data as learning data includes the time when the patient received treatment, a treatment history indicating a combination of multiple types of treatments the patient received, and a condition history indicating the patient's condition at that time. At this time, the information processing device 10 acquires patient data for each patient at multiple times as learning data, and also acquires patient data for multiple patients.

[0027] Next, the information processing device 10 performs a process of classifying the treatment history A included in the patient data into one of a plurality of preset clusters (step S2). For example, the information processing device 10 first converts the treatment history A at each time t into a treatment history vector AV expressed in a vector representation with the type of treatment as an element, as shown in Fig. 3. Then, the information processing device 10 classifies the treatment history vector AV into a corresponding cluster C according to the characteristics of the vector array, as shown in Fig. 4.

[0028] Next, the information processing device 10 performs machine learning using the condition history X of each patient and each time and the cluster C to generate a clustering model M1 (step S3), as shown in Fig. 5. As a result, the clustering model M1 is trained to output a new cluster in response to the input of a new condition history of the patient. Then, the information processing device 10 stores the generated clustering model M1 in the model storage unit 17.

[0029] Next, as shown in FIG. 6, the information processing device 10 performs machine learning using the condition history X for each patient and each time point, the treatment history vector AV of the treatment history X, the cluster C into which the treatment history X is classified, and the cluster centroid vector CV representing the characteristics of the cluster C as a vector with multiple types of treatment as elements, to generate a restoration model M2 (step S4). At this time, the cluster C used in the machine learning only requires identification information to identify the cluster, and the cluster centroid vector CV previously associated with the cluster identification information is used in the machine learning. As a result, the restoration model M2 is trained to output a new treatment proposal in response to the input of a new condition history of the patient and the cluster centroid vector corresponding to the cluster output by inputting the new condition history into the clustering model M1. The information processing device 10 then stores the generated restoration model M2 in the model storage unit 17.

[0030] Next, the operation of the information processing device 10 when selecting a treatment method will be described with reference to Fig. 11. The information processing device 10 acquires current patient data from the patient as selection data (step S11). The patient data as selection data is a condition history that indicates the condition of the patient at the current time t, as shown in Fig. 8.

[0031] Next, the information processing device 10 reads out the clustering model M1 and inputs the patient's condition history to the clustering model M1. Then, as shown in Fig. 8, a new cluster C corresponding to the input condition history is output from the clustering model M1, and the information processing device 10 acquires the new cluster C (step S12).

[0032] Next, the information processing device 10 reads out the restoration model M2 and inputs to the restoration model M2 the patient's condition history and information based on the new cluster C output from the clustering model M1. Specifically, as shown in Fig. 9, the information processing device 10 inputs to the restoration model the cluster centroid vector associated with the identification information of the new cluster C and the patient's condition history. Then, the restoration model M2 outputs a treatment proposal B which is a vector representation of a combination of new treatments corresponding to the input condition history and cluster centroid vector, and the information processing device 10 acquires the new treatment proposal B (step S13).

[0033] The information processing device 10 then extracts multiple treatments to be performed from the new treatment proposal B, which is composed of the acquired vector representation, and outputs them to a display screen or the like. This allows a person who has been shown the treatment proposal to refer to the treatment proposal and set a treatment method. As a result, when treating a patient, a treatment proposal that combines multiple treatments according to the condition of each individual patient can be easily made.

[0034] <Embodiment 2> Next, a second embodiment of the present invention will be described with reference to Fig. 12 and Fig. 13. Fig. 12 and Fig. 13 are diagrams for explaining the processing operation of the information processing device 10 in the second embodiment.

[0035] The information processing device 10 in this embodiment has a configuration similar to that shown in FIG. 1 and described in the first embodiment. However, in this embodiment, the data structure of the learned treatment history and the treatment proposal output from the model differs from that in the first embodiment. Furthermore, in this embodiment, "treatment" is replaced with "menu" and "patient" with "subject," and an example will be described in which a combination of multiple menus used in the fields of rehabilitation and training is learned and proposed. However, the "menu" described in this embodiment may be the "treatment" described in the first embodiment, or any "measure" (e.g., action, proposal) that can be implemented depending on the subject's condition. Below, the configuration of the information processing device 10 that differs from that in the first embodiment will be mainly described.

[0036] During model generation, the input unit 11 receives and stores, as learning data, subject data including the time (e.g., date) when the subject performed a menu, a menu history (combination information) representing a combination of multiple menus (measures) performed by the subject, and a status history (status information) representing the subject's status at that time. In this embodiment, as shown in FIG. 12 , a higher-level menu (first measure) set in a higher hierarchy (first hierarchy) and a lower-level menu (second measure) set in a lower hierarchy (second hierarchy) are set. A higher-level menu in a higher hierarchy and a lower-level menu in a lower hierarchy are paired to form a single menu. A combination of multiple paired menus constitutes a menu history performed at a given time. For example, a higher-level menu set in a higher hierarchy may represent exercise content such as "swimming" or "jogging," while a lower-level menu set in a lower hierarchy may represent time content such as an exercise duration such as "one hour" or an exercise timing such as "morning." These paired menus constitute a single menu (e.g., "swimming" for "one hour"), and a menu history is formed by combining multiple paired menus. In the example of Figure 12, at time t1, the menu history is composed of two pairs of menus: "a pair of upper menu a and lower menu c'" and "a pair of upper menu z and lower menu z'".

[0037] The clustering unit 12 performs a process of classifying the menu history, which is the above-described learning data, into one of a plurality of preset clusters. At this time, the clustering unit 12 first separates the menu history into upper menus and lower menus, and converts each into a vector representation. For example, as shown in FIG. 12, the menu history is separated into an upper menu vector AV1 (first combination information) consisting of upper menus in an upper hierarchy and a lower menu vector AV1' (second combination information) consisting of lower menus in a lower hierarchy. At this time, the upper menu vector AV1 and the lower menu vector AV1' are associated with each other as a pair of upper menus and lower menus. The clustering unit 12 then further classifies the upper menu vector AV1 separated into the upper hierarchy into one of the preset upper hierarchy clusters C (first clusters) according to the characteristics of the vector arrangement. Similarly, the clustering unit 12 classifies the lower menu vector AV1' separated into the lower hierarchy into one of the preset lower hierarchy clusters C (second clusters) according to the characteristics of the vector arrangement. This processing is performed for each upper layer and each lower layer by the method described in the first embodiment with reference to FIG.

[0038] The first learning unit 13 (first model generation unit) generates clustering models M11 and M12 (first models) for each upper hierarchical level and each lower hierarchical level, respectively, in the same manner as in the first embodiment. Specifically, the first learning unit 13 inputs the subject's state history X, the subject's upper menu vector AV1, and the upper hierarchical cluster C at each time t for learning data classified in the upper hierarchical level, thereby learning the relationships among them and generating a clustering model M11 for the upper hierarchical level that outputs a new upper hierarchical cluster in response to the input of new state history. Similarly, the first learning unit 13 inputs the subject's state history X, the subject's lower menu vector AV1′, and the lower hierarchical cluster C at each time t for learning data classified in the lower hierarchical level, thereby learning the relationships among them and generating a clustering model M12 for the lower hierarchical level that outputs a new lower hierarchical cluster in response to the input of new state history. Note that this processing is performed for each upper hierarchical level and each lower hierarchical level using the method described with reference to FIG. 5 in the first embodiment. In this way, the first learning unit 13 generates a clustering model M11 corresponding to the upper layer and a clustering model M12 corresponding to the lower layer.

[0039] The second learning unit 14 (second model generation unit) performs machine learning using the subject's state history for each time period, the upper menus in the menu history, the upper layer clusters into which the upper menus are classified, the lower menus in the menu history, and the lower layer clusters into which the lower menus are classified, to generate a restoration model M20 (second model) that outputs new menu suggestions. Specifically, in this embodiment, the second learning unit 14 performs machine learning using, in addition to the above information, an upper cluster centroid vector associated with the identification information of the upper layer cluster and representing the features of the upper layer cluster as a vector with each upper menu as an element, and a lower cluster centroid vector associated with the identification information of the lower layer cluster and representing the features of the lower layer cluster as a vector with each lower menu as an element. Furthermore, the second learning unit 14 performs machine learning using, in addition to the above information, an upper menu vector AV1 and a lower menu vector AV1' that represent the menu history, with the paired upper menu and lower menu associated with each other. In this way, the second learning unit 14 receives as input the state history, the identification information and upper cluster centroid vector of the upper hierarchical cluster, the identification information and lower cluster centroid vector of the lower hierarchical cluster, and the upper menu vector AV1 and the lower menu vector AV1' in which the paired upper menu and lower menu are associated, and machine-learns the relationship between them. As a result, the second learning unit 14 generates a restoration model M20 that outputs new menu suggestions in response to inputs of the subject's new state history, the upper cluster centroid vector corresponding to the upper hierarchical cluster output by inputting the new state history to the upper clustering model M11, and the lower cluster centroid vector corresponding to the lower hierarchical cluster output by inputting the new state history to the lower clustering model M12.

[0040] The above-described machine learning by the second learning unit 14 is performed in the same manner as the method described in embodiment 1 with reference to Fig. 6. However, the method according to this embodiment differs from the method shown in Fig. 6 in that clusters and cluster centroid vectors for the upper and lower hierarchical levels are input, and that the menu history includes upper menu vectors and lower vector menus, and paired upper and lower menus are input in an associated state.

[0041] Furthermore, when performing a process of selecting a future menu for the subject, the input unit 11 accepts input of a state history (state information) that indicates the state of the subject at that time as selection data.

[0042] The output unit 15 then reads out the upper layer clustering model M11, the lower layer clustering model M12, and the restored model M20 generated as described above, and performs a process of selecting a future menu to be proposed to the subject from the subject's state history accepted as selection data. Specifically, as shown in Fig. 13, the output unit 15 first inputs the state history to the upper layer clustering model M11 and the lower layer clustering model M12. Then, the upper layer clustering model M11 outputs a new upper layer cluster corresponding to the input state history, and the lower layer clustering model M12 outputs a new lower layer cluster corresponding to the input state history, and the output unit 15 acquires the new upper layer cluster and lower layer cluster, respectively.

[0043] Next, the output unit 15 inputs the state history, information based on the new upper cluster, and information based on the new lower cluster to the restored model M20, as shown in Fig. 13. At this time, specifically, similar to the process described with reference to Fig. 9 in the first embodiment, the output unit 15 inputs the upper cluster centroid vector associated with the identification information of the new upper cluster, the lower cluster centroid vector associated with the identification information of the new lower cluster, and the state history to the restored model M20. Then, as shown in Fig. 13, the restored model M20 outputs menu proposal B, which represents, in vector form, combinations of new pairs of upper menus and lower menus corresponding to the input state history, upper cluster centroid vector, and lower cluster centroid vector, and the output unit 15 acquires this.

[0044] In this way, even with a combination of hierarchical menus (measures) as in this embodiment, it is possible to easily make proposals that combine multiple menus (measures) according to the condition of each individual subject.

[0045] <Embodiment 3> Next, a third embodiment of the present invention will be described with reference to Fig. 14 to Fig. 16. Fig. 14 to Fig. 15 are block diagrams showing the configuration of an information processing device in the third embodiment, and Fig. 16 is a flowchart showing the operation of the information processing device. Note that this embodiment shows an outline of the configuration of the information processing device and information processing method described in the above embodiments.

[0046] First, the hardware configuration of the information processing device 100 in this embodiment will be described with reference to Fig. 14. The information processing device 100 is configured as a general information processing device, and is equipped with the following hardware configuration, as an example. ·CPU(Central Processing Unit)101(Arithmetic unit) ROM (Read Only Memory) 102 (storage device) RAM (Random Access Memory) 103 (storage device) Programs 104 loaded into RAM 103 A storage device 105 for storing a group of programs 104 A drive device 106 that reads and writes from a storage medium 110 external to the information processing device A communication interface 107 that connects to a communication network 111 outside the information processing device Input / output interface 108 for inputting and outputting data Bus 109 connecting each component

[0047] The information processing device 100 can construct and include the clustering 121, first model generation unit 122, and second model generation unit 123 shown in FIG. 15 by having the CPU 101 acquire and execute the program group 104. The program group 104 is stored in advance in, for example, the storage device 105 or the ROM 102, and is loaded into the RAM 103 and executed by the CPU 101 as needed. The program group 104 may be supplied to the CPU 101 via the communication network 111, or may be stored in advance in the storage medium 110, and the drive device 106 may read out the programs and supply them to the CPU 101. However, the clustering 121, first model generation unit 122, and second model generation unit 123 described above may be constructed using dedicated electronic circuits for realizing such means.

[0048] 14 shows an example of the hardware configuration of the information processing device 100, and the hardware configuration of the information processing device is not limited to the above-described case. For example, the information processing device may be configured with only a part of the above-described configuration, such as excluding the drive device 106.

[0049] Then, the information processing device 100 executes the information processing method shown in the flowchart of FIG. 16 by the functions of the clustering 121, the first model generation unit 122, and the second model generation unit 123 constructed by the program as described above.

[0050] As shown in FIG. 16, the information processing device 100 Combination information representing a combination of multiple types of measures implemented on a target person at each time point is classified into one of multiple preset clusters (step S101); Based on state information representing the state of the target person at each time point and the clusters into which the multiple types of measures implemented by the target person are classified, a first model is generated that outputs the clusters for the state information (step S102); generating a second model that outputs the combination information for the state information and the information based on the clusters, based on the state information, the clusters, and the combination information for each time period (step S103); Execute the process.

[0051] With the above-described configuration, the present invention can output new combination information of multiple measures based on the new state of the target person by using the generated first and second models. For example, when treating a patient, it is possible to easily propose a treatment that combines multiple treatments depending on the state of each individual patient.

[0052] The above-described program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can supply the program to a computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

[0053] Although the present invention has been described above with reference to the above-described embodiments, the present invention is not limited to the above-described embodiments. Various modifications that are understandable to those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. Furthermore, at least one or more of the functions of the clustering unit 121, the first model generation unit 122, and the second model generation unit 123 described above may be executed by an information processing device installed and connected anywhere on a network, i.e., may be executed by so-called cloud computing.

[0054] <Additional Notes> A part or all of the above-described embodiments can be described as follows: The following provides an overview of the configurations of the information processing method, information processing device, and program according to the present invention. However, the present invention is not limited to the following configurations. (Appendix 1) Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Information processing methods. (Appendix 2) 10. The information processing method according to claim 1, classifying the combination information into any of the clusters, the number of which is set to be less than the number of types of measures; Information processing methods. (Appendix 3) 10. The information processing method according to claim 1 or 2, classifying the combination information of vector expressions with the types of measures as elements into one of the clusters according to the characteristics of the vector expressions; Information processing methods. (Appendix 4) An information processing method according to any one of Supplementary Notes 1 to 3, generating the second model that outputs the combination information for the state information and the cluster feature information based on the state information, the cluster, the combination information, and cluster feature information representing features of the cluster for each time period; Information processing methods. (Appendix 5) 4. The information processing method according to claim 3, generating the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the cluster, the combination information in vector representation for each time period, and cluster characteristic information in vector representation having as elements the type of measures that represent the characteristics of the cluster; Information processing methods. (Appendix 6) 6. An information processing method according to any one of Supplementary Notes 1 to 5, When a type of measure is configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, a combination of a plurality of types of paired measures implemented on a target person at each time point is separated into first combination information representing a combination of first measures belonging to the first hierarchical level and second combination information representing a combination of second measures belonging to the second hierarchical level, and the first combination information is classified into one of a plurality of first clusters set in advance, and the second combination information is classified into one of a plurality of second clusters set in advance, generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; Information processing methods. (Appendix 7) 10. The information processing method according to claim 6, generating the second model that outputs a plurality of combinations of paired measures for the state information, information based on the first cluster, and information based on the second cluster, based on the state information, the first cluster, the second cluster, and a plurality of combinations of paired first measures and second measures for each time period; Information processing methods. (Appendix 8) 8. An information processing method according to any one of Supplementary Notes 1 to 7, inputting new state information into the first model to output new clusters; and outputting new combination information by inputting the new cluster-based information output from the first model and the new state information into the second model. Information processing methods. (Appendix 9) 6. The information processing method according to claim 4 or 5, inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; Information processing methods. (Appendix 10) a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; An information processing device comprising: (Appendix 11) 11. The information processing device according to claim 10, the clustering unit classifies the combination information into any of the clusters, the number of which is set to be less than the number of types of measures. Information processing device. (Appendix 12) 12. The information processing device according to claim 10, the clustering unit classifies the combination information of vector expressions having types of measures as elements into one of the clusters according to characteristics of the vector expressions. Information processing device. (Appendix 13) 13. The information processing device according to any one of Supplementary Notes 10 to 12, the second model generation unit generates the second model that outputs the combination information for the state information and the cluster feature information based on the state information, the clusters, the combination information for each time period, and cluster feature information that represents features of each cluster for a plurality of types of measures. Information processing device. (Appendix 14) 13. The information processing device according to claim 12, the second model generation unit generates the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the cluster, the combination information in vector representation for each time period, and also on cluster characteristic information in vector representation having types of measures representing characteristics of the cluster as elements. Information processing device. (Appendix 15) 15. The information processing device according to any one of Supplementary Notes 10 to 14, when a type of measure is configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, the clustering unit separates combinations of multiple types of paired measures implemented on a target person for each time period into first combination information representing combinations of first measures belonging to the first hierarchical level and second combination information representing combinations of second measures belonging to the second hierarchical level, classifies the first combination information into one of a plurality of first clusters set in advance, and classifies the second combination information into one of a plurality of second clusters set in advance; The first model generation unit generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; The second model generation unit generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; Information processing device. (Appendix 16) 16. The information processing device according to claim 15, The second model generation unit generating the second model that outputs a plurality of combinations of paired measures for the state information, information based on the first cluster, and information based on the second cluster, based on the state information, the first cluster, the second cluster, and a plurality of combinations of paired first measures and second measures for each time period; Information processing device. (Appendix 17) 17. The information processing device according to any one of Supplementary Notes 10 to 16, inputting new state information into the first model to output new clusters; An information processing device including an output unit that outputs new combination information by inputting information based on the new cluster output from the first model and the new state information to the second model. (Appendix 18) 15. The information processing device according to claim 13, inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; An information processing device having an output unit. (Appendix 19) In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; A computer-readable storage medium that stores a program for executing a process. [Explanation of symbols]

[0055] 10. Information processing equipment 11 Input section 12 Clustering Department 13 First Learning Department 14 Second Learning Department 15 Output section 16 Data storage unit 17 Model memory section 20 Data management device 100 Information processing device 101 CPU 102 ROM 103 RAM 104 Programs 105 Storage device 106 Drive device 107 Communication Interface 108 Input / Output Interface 109 Bus 110 Storage medium 111 Communication Network 121 Clustering Department 122 First model generation unit 123 Second Model Generation Unit

Claims

1. An information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Furthermore, when the types of measures are configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, the combinations of multiple types of paired measures implemented on the target person at each time point are separated into first combination information representing a combination of first measures belonging to the first hierarchical level and second combination information representing a combination of second measures belonging to the second hierarchical level, and the first combination information is classified into one of a plurality of first clusters set in advance, and the second combination information is classified into one of a plurality of second clusters set in advance, generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; Information processing methods.

2. An information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Furthermore, by inputting new state information into the first model, new clusters are output; and outputting new combination information by inputting the new cluster-based information output from the first model and the new state information into the second model. Information processing methods.

3. An information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; further generating the second model that outputs the combination information for the state information and the cluster feature information based on cluster feature information that represents features of the cluster in addition to the state information, the cluster, and the combination information for each time period; inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; Information processing methods.

4. An information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies combinations of multiple types of measures implemented on the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Furthermore, the combination information of the vector expression having the type of measures as an element is classified into one of the clusters according to the characteristics of the vector expression; generating the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the cluster, the combination information expressed in vector form for each time period, and cluster characteristic information expressed in vector form, the cluster characteristic information having as elements a type of measure that represents a characteristic of the cluster; inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; Information processing methods.

5. 5. An information processing method according to claim 1, The information processing device, classifying the combination information into any of the clusters, the number of which is set to be less than the number of types of measures; Information processing methods.

6. 4. An information processing method according to claim 1, The information processing device, classifying the combination information of vector expressions with the types of measures as elements into one of the clusters according to characteristics of the vector expressions; Information processing methods.

7. 3. The information processing method according to claim 1 or 2, The information processing device, generating the second model that outputs the combination information for the state information and the cluster feature information based on the state information, the cluster, the combination information, and cluster feature information representing features of the cluster for each time period; Information processing methods.

8. 3. The information processing method according to claim 2, The information processing device, When a type of measure is configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, a combination of a plurality of types of paired measures implemented on a target person at each time point is separated into first combination information representing a combination of first measures belonging to the first hierarchical level and second combination information representing a combination of second measures belonging to the second hierarchical level, and the first combination information is classified into one of a plurality of first clusters set in advance, and the second combination information is classified into one of a plurality of second clusters set in advance, generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; Information processing methods.

9. 9. The information processing method according to claim 1 or 8, The information processing device, generating the second model that outputs a plurality of combinations of paired measures for the state information, information based on the first cluster, and information based on the second cluster, based on the state information, the first cluster, the second cluster, and a plurality of combinations of paired first measures and second measures for each time period; Information processing methods.

10. a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; Equipped with when a type of measure is configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, the clustering unit separates combinations of multiple types of paired measures implemented on a target person for each time period into first combination information representing combinations of first measures belonging to the first hierarchical level and second combination information representing combinations of second measures belonging to the second hierarchical level, classifies the first combination information into one of a plurality of first clusters set in advance, and classifies the second combination information into one of a plurality of second clusters set in advance; The first model generation unit generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; The second model generation unit generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; Information processing device.

11. a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; inputting new state information into the first model to output new clusters; and outputting new combination information by inputting the new cluster-based information output from the first model and the new state information into the second model. an output unit; An information processing device comprising:

12. a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; Equipped with the second model generation unit generates the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the clusters, the combination information, and cluster characteristic information that represents characteristics of each of the clusters for a plurality of types of measures, for each time period; Furthermore, by inputting new state information into the first model, new clusters are output; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; an output section; Information processing device.

13. a clustering unit that classifies combination information representing a combination of multiple types of measures implemented on a target person for each time period into one of multiple preset clusters; a first model generation unit that generates a first model based on status information representing the status of a target person at each time point and the clusters that classify combinations of multiple types of measures implemented on the target person, and outputs the clusters for the status information; a second model generation unit that generates a second model based on the state information, the cluster, and the combination information for each time period, the second model outputting the combination information for information based on the state information and the cluster; Equipped with the clustering unit classifies the combination information of vector expressions having types of measures as elements into one of the clusters according to characteristics of the vector expressions; the second model generation unit generates the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the cluster, the combination information expressed in vectors for each time period, and cluster characteristic information expressed in vectors, the elements of which are types of measures that represent characteristics of the clusters; Furthermore, by inputting new state information into the first model, new clusters are output; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; an output section; Information processing device.

14. In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Execute the process, Furthermore, when the types of measures are configured by a pair of a first measure belonging to a first hierarchical level and a second measure belonging to a second hierarchical level, the combinations of multiple types of paired measures implemented on the target person at each time point are separated into first combination information representing a combination of first measures belonging to the first hierarchical level and second combination information representing a combination of second measures belonging to the second hierarchical level, and the first combination information is classified into one of a plurality of first clusters set in advance, and the second combination information is classified into one of a plurality of second clusters set in advance, generating the first model corresponding to a first hierarchical level that outputs the first cluster for the state information based on state information that represents the state of the target person at each time point and the first cluster that classifies the measures implemented for the target person; generating the first model corresponding to a second hierarchical level that outputs the second cluster for the state information based on state information that represents the state of the target person at each time point and the second cluster that classifies the measures implemented on the target person; generating the second model that outputs the first combination information and the second combination information for the state information, the information based on the first cluster, and the information based on the second cluster, based on the state information, the first cluster, the second cluster, the first combination information, and the second combination information for each time period; A program for executing a process.

15. In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Execute the process, Furthermore, by inputting new state information into the first model, new clusters are output; and outputting new combination information by inputting the new cluster-based information output from the first model and the new state information into the second model. A program for executing a process.

16. In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Execute the process, further generating the second model that outputs the combination information for the state information and the cluster feature information based on cluster feature information that represents features of the cluster in addition to the state information, the cluster, and the combination information for each time period; inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; A program for executing a process.

17. In the information processing device, Classifying combination information representing a combination of multiple types of measures implemented on the target person for each time period into one of multiple preset clusters; generating a first model that outputs the cluster for the state information based on state information that represents the state of the target person at each time point and the cluster that classifies multiple types of measures implemented by the target person; generating a second model that outputs the combination information for the state information and the information based on the cluster, based on the state information, the cluster, and the combination information for each time period; Execute the process, Furthermore, the combination information of the vector expression having the type of measures as an element is classified into one of the clusters according to the characteristics of the vector expression; generating the second model that outputs the combination information for the state information and the cluster characteristic information based on the state information, the cluster, the combination information expressed in vector form for each time period, and cluster characteristic information expressed in vector form, the cluster characteristic information having as elements a type of measure that represents a characteristic of the cluster; inputting new state information into the first model to output new clusters; outputting new combination information by inputting the cluster feature information of the new cluster output from the first model and the new state information into the second model; A program for executing a process.

Images