Population Status Determination System

Abstract

The present invention suitably determines the state of a population.　A population state determination system 10 comprises: an acquisition unit 11 that acquires population information indicating a population in a time series for an area in which the state of the population is to be determined; a model calculation unit 12 that inputs the population information acquired by the acquisition unit 11 into an encoder-decoder model, which is stored in advance and which compresses and restores input data, and that performs a calculation, obtaining an output from the encoder-decoder model; a determination unit 13 that compares the population information acquired by the acquisition unit 11 and the output obtained by the model calculation unit 12 and determines the state of the population in the area; and a determination criteria generation unit 14 that generates determination criteria used for determination by the determination unit 13. The determination criteria generation unit 14 generates the determination criteria by inputting population information for determination criteria generation to an encoder-decoder model for determination criteria generation that is stored in advance and then performing calculation.

Description

[Technical Field]

[0001] The present invention relates to a population status determination system for determining the population status of an area. [Background technology]

[0002] BACKGROUND ART Conventionally, a technique has been proposed for estimating the population for each area and time period using data from mobile terminals such as mobile phones (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-123011 Summary of the Invention [Problem to be solved by the invention]

[0004] Using the estimated population information, it is possible to detect areas with abnormal populations, such as areas with higher than normal populations. This makes it possible to detect sudden events and discover places where many people gather during disasters.

[0005] One method for detecting population anomalies is a statistical method based on the average and variance of population in an area and time period. This method can detect anomalies in a certain area and time period. However, this method does not take population trends into account, and therefore cannot necessarily detect anomalies with high accuracy.

[0006] An embodiment of the present invention has been made in view of the above, and aims to provide a population status determination system that can appropriately determine the population status. [Means for solving the problem]

[0007] In order to achieve the above object, a population status determination system according to one embodiment of the present invention comprises: an acquisition unit that acquires population information indicating the time-series population of an area whose population status is to be determined; a model calculation unit that inputs the population information acquired by the acquisition unit into a pre-stored encoder-decoder model that compresses and decompresses input data, performs calculations to obtain output from the encoder-decoder model; a determination unit that compares the population information acquired by the acquisition unit with the output obtained by the model calculation unit to determine the population status of the area; and a determination criterion generation unit that generates a determination criterion to be used in the determination by the determination unit. The determination criterion generation unit acquires population information for generating the determination criterion, including first population information for generating the determination criterion in a first state of the same area and a second state different from the first state, and second population information for generating the determination criterion in the first state of the area to be determined, inputs the acquired first and second population information into a pre-stored encoder-decoder model for generating the determination criterion, performs calculations to obtain output from the encoder-decoder model, compares the input and output of the encoder-decoder model, and generates a determination criterion based on the comparison result.

[0008] According to a population status determination system of one embodiment of the present invention, it is possible to determine the population status taking into account the time-series population of an area. Furthermore, the input to the encoder-decoder model is compared with the output to make a determination. Furthermore, an appropriate determination criterion is generated based on first and second population information for generating the determination criterion, and is used for the determination. Therefore, according to the population status determination system of one embodiment of the present invention, it is possible to appropriately determine the population status. [Effects of the Invention]

[0009] According to one embodiment of the present invention, the state of the population can be appropriately determined. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing the configuration of a computer that is a population status determination system and a model generation system according to an embodiment of the present invention. [Figure 2] 10 is a graph showing an example of population information and output values ​​from an encoder-decoder model when the population information is used as an input value. [Figure 3] FIG. 10 is a diagram illustrating an example of information used in a computer. [Figure 4] FIG. 1 is a diagram illustrating an example of an encoder-decoder model that can be generated and used by a computer. [Figure 5] FIG. 10 is a diagram illustrating another example of an encoder-decoder model that can be generated and used by a computer. [Figure 6] FIG. 10 is a diagram illustrating an example of information used in a computer. [Figure 7] FIG. 2 is a diagram schematically illustrating training examples and test examples, which are first and second population information for generating a criterion used to generate a criterion. [Figure 8] 1 is a flowchart illustrating processing executed in a model generation system according to an embodiment of the present invention. [Figure 9] 1 is a flowchart showing a process executed in a population status determination system according to an embodiment of the present invention. [Figure 10] 10 is a flowchart showing a process executed when generating a determination criterion in the population status determination system according to the embodiment of the present invention. [Figure 11] FIG. 1 is a diagram illustrating the hardware configuration of a computer that is a population status determination system and a model generation system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of the population status determination system according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are given the same reference numerals, and duplicated explanations will be omitted.

[0012] FIG. 1 shows a computer 1 that is a population status determination system 10 and a model generation system 20 according to this embodiment. The population status determination system 10 is a system (device) that determines (estimates) the population status of a geographical area. The area to be determined is, for example, a 500m square area divided into regions. A half regional mesh may be used as the area. Alternatively, administrative divisions such as cities, towns, villages, or prefectures, or pre-set land use divisions may be used as the area. In the following explanation, the area will be described as a mesh. Note that the area to be determined does not have to be the above and can be any geographical area.

[0013] The determination by the population status determination system 10 is based on population information indicating the time series population of the area to be determined. For example, as described below, population information indicating the population for each hour of a day is used for this determination. This determination is, for example, a determination of whether the population in the area to be determined is in an abnormal state that differs from normal. That is, this determination detects an abnormality in the population trend in the area to be determined. An abnormal state in which the population is in a state that differs from normal is, for example, a state in which the population trend differs excessively from normal population trends. This determination can, for example, detect sudden events or discover locations where large numbers of people are congregating during a disaster. Note that the determination by the population status determination system 10 may determine the degree of abnormality rather than whether an abnormal state is present. Alternatively, the determination by the population status determination system 10 may be any other determination of the population status of an area.

[0014] As will be described later, the population status determination system 10 performs a determination by performing calculations on population information using an encoder-decoder model, which is a trained model generated by machine learning. The encoder-decoder model compresses and decompresses input data. The model generation system 20 generates the encoder-decoder model used for determination by the population status determination system 10.

[0015] A conventional computer can be used as the computer 1 that is the population status determination system 10 and the model generation system 20 according to this embodiment. The computer 1 may also be a computer system including multiple computers.

[0016] Next, the functions of the population status determination system 10 and the model generation system 20 according to this embodiment will be described. First, the functions of the model generation system 20 will be described, followed by the functions of the population status determination system 10. As shown in FIG. 1 , the model generation system 20 is configured to include a learning acquisition unit 21 and a model generation unit 22.

[0017] The learning acquisition unit 21 is a functional unit that acquires learning population information indicating a time-series population, which is used to generate an encoder-decoder model. The learning acquisition unit 21 may acquire learning type information indicating the type of area related to the learning population information. The learning acquisition unit 21 may acquire the learning type information by performing clustering using the learning population information. The learning acquisition unit 21 acquires each piece of information as follows.

[0018] Each piece of learning population information is in the same format as the population information used to determine the population status. For example, the population information is information showing the population of an area for each hour of a day (midnight, 1 a.m., ..., 11 p.m.). Figure 2 shows a portion of a graph G1, an example of population information. When such population information is used, the population status determination system 10 determines the population status of the area being determined on that day. Note that the overall time period (one day in the above example), time interval (every hour in the above example), and format of the population information to be determined do not necessarily have to be as described above.

[0019] A large amount of training population information is used to generate an encoder-decoder model. The large amount of training population information typically includes training population information for multiple areas. The training acquisition unit 21 acquires, for example, the data shown in FIG. 3(a). The data shown in FIG. 3(a) is information in which a mesh code (information in the "meshcode" column), information indicating a time (information in the "timestamp" column), and information indicating a population (information in the "population" column) are associated with each other. The mesh code is information such as a character string that identifies a mesh, which is an area, and is set in advance for each area. The information indicating a time is, for example, information indicating the year, month, date, and time of day. The information indicating a population indicates the population in the area and at the time indicated by the corresponding mesh code and information indicating the time.

[0020] The population data shown in Figure 3(a) is generated as spatial statistical information from information indicating the location of mobile phones and registered information about mobile phone subscribers using an existing method, for example. The population data shown in Figure 3(a) may also be generated using any method other than the above. The learning acquisition unit 21 acquires the population data shown in Figure 3(a) that is pre-stored in a database of the computer 1 or another device.

[0021] The learning acquisition unit 21 formats the acquired data into data for each mesh code and for each hour of the day (0:00, 1:00, ..., 23:00) as shown in Figure 3(b), i.e., data on population trends for each day on an area-by-area basis. This population trend data corresponds to learning population information. The learning acquisition unit 21 acquires a sufficient amount of population trend data to generate an encoder-decoder model using machine learning. The population trend data may or may not include data on the area for which the population status is to be determined. Note that the learning acquisition unit 21 may also acquire information indicating a time series population other than the above as learning population information.

[0022] The learning acquisition unit 21 may acquire learning type information indicating the type of area related to the population transition data. The area type is a type that may affect the population transition in the area. For example, the area type may be an urban type such as "business district" or "residential district."

[0023] For example, the learning acquisition unit 21 acquires learning type information that is pre-stored in a database of the computer 1 or another device. FIG. 3(c) shows an example of data that is pre-stored learning type information. The data shown in FIG. 3(c) is information in which a mesh code (information in the "meshcode" column), information indicating a city type (information in the "city type" column), and a type code (information in the "type code" column) are associated with each other. The information indicating a city type is information that indicates the meaning of the type of area indicated by the corresponding mesh code. The information indicating a city type is set in advance for each area. Note that the information indicating a city type does not need to be used for processing in the model generation system 20, and therefore does not need to be acquired.

[0024] The type code is information that identifies the type of area indicated by the corresponding mesh code (a flag indicating the area), and is set in advance for each area. The type code is a numerical value that can be used for machine learning. The type code is the same numerical value for the same city type, and different numerical values ​​for different city types. The learning acquisition unit 21 acquires the type code corresponding to the mesh code of the area related to the population transition data as learning type information.

[0025] The learning acquisition unit 21 may acquire the learning type information by performing clustering using the learning population information, rather than acquiring pre-stored learning type information. The learning acquisition unit 21 performs clustering using the data of population changes per day for each area. For example, the learning acquisition unit 21 performs area clustering using the data of population changes per day for each area, as follows. By performing such clustering, areas with similar population changes can be divided into clusters.

[0026] The learning acquisition unit 21 takes the average of the population for each time period in each area, and generates one population transition data for each area. For example, the learning acquisition unit 21 takes the average for each time period of the population transition data for each day for a period set in advance for each area (for example, the period from one month before the present time to the present time), and generates one population transition data for each area. The learning acquisition unit 21 clusters the population transition data to cluster the areas. The clustering itself may be performed by a conventional method (for example, the k-means method).

[0027] Alternatively, the learning acquisition unit 21 may cluster the population transition data that may include multiple population transition data for one area. For each area, the learning acquisition unit 21 determines the cluster that includes the most population transition data as the cluster for that area.

[0028] The learning acquisition unit 21 assigns a different type code (cluster number) to each cluster. The learning acquisition unit 21 uses the type code of the cluster to which an area belongs as learning type information for that area. The learning acquisition unit 21 stores the correspondence between the mesh code and type code for each area in the computer 1 so that it can also be used in the population status determination system 10. Note that when learning type information is acquired by performing clustering, there is no information indicating the city type.

[0029] The learning acquisition unit 21 outputs the acquired learning population information to the model generation unit 22. In addition, in an aspect in which the learning acquisition unit 21 acquires learning type information, the learning acquisition unit 21 also outputs the acquired learning type information to the model generation unit 22.

[0030] The model generation unit 22 is a functional unit that performs machine learning based on the learning population information acquired by the learning acquisition unit 21, and generates an encoder-decoder model that inputs information indicating a time-series population. The model generation unit 22 may generate an encoder-decoder model based on the learning type information acquired by the learning acquisition unit 21. The model generation unit 22 may generate an encoder-decoder model that also inputs type information indicating an area type. The model generation unit 22 may generate multiple encoder-decoder models according to type information indicating an area type.

[0031] The encoder-decoder model generated by the model generation unit 22 is explained below. Figure 4 shows an example of an encoder-decoder model. The encoder-decoder model is composed of a neural network and is trained to input population information indicating the time series population of an area, perform dimensionality reduction, and output the original population information. Examples of encoder-decoder models include autoencoders (Geoffrey Hinton and Salakhutdinov Ruslan, “Reducing the dimensionality of data with neural networks,” Science, pp. 504-507, 2006) and Transformers (Ashish Vaswani et al., “Attention Is All You Need,” Advances in neural information processing systems, 2017). The input layer of the encoder-decoder model has neurons equal to the number of elements in the population information (the dimensionality of the population information). If the population information is numerical information indicating the population of an area for each hour of the day (midnight, 1:00, ..., 23:00), the input layer of the encoder-decoder model has 24 neurons (vectors) that input the population values ​​for each hour of the day. The output layer of the encoder-decoder model has the same number of neurons (vectors) as the input layer, each corresponding to a neuron in the input layer.

[0032] The configuration of the encoder-decoder model itself may be the same as that of a conventional encoder-decoder model. As shown in Figure 4, a hidden layer with multiple neurons (vectors) is provided between the input layer and the output layer. Each neuron in the input layer and each neuron in the hidden layer are connected with a weight w used for calculation. Furthermore, each neuron in the hidden layer and each neuron in the output layer are connected with a weight w used for calculation. The number of neurons in the hidden layer is less than the number of neurons in the input layer and the output layer. This allows for dimensionality reduction in the hidden layer.

[0033] The model generation unit 22 generates an encoder-decoder model as follows: First, an example of a mode in which learning type information is not used will be described, and then an example of a mode in which learning type information is used will be described.

[0034] The model generation unit 22 receives population transition data, which is the learning population information, from the learning acquisition unit 21. As shown in FIG. 4, the model generation unit 22 performs machine learning using the population transition data as the input value to the encoder-decoder model and the output value (correct answer) of the encoder-decoder model to generate an encoder-decoder model. The above machine learning to generate the encoder-decoder model can be performed in the same way as conventional machine learning methods. The above is an example in which learning type information is not used.

[0035] Next, an example of a mode in which the learning type information is used will be described. The model generation unit 22 inputs the type code, which is the learning type information, together with the population transition data from the learning acquisition unit 21. In this case, the model generation unit 22 generates an encoder-decoder model that also inputs the type code. Figure 5 shows an example of this encoder-decoder model. In addition to the encoder-decoder model shown in Figure 4, this encoder-decoder model has neurons corresponding to the type code in the input layer and output layer.

[0036] As shown in Fig. 6(a), the model generation unit 22 associates the area type code with the area and daily population transition data. This association is performed using the mesh code as a key. As shown in Fig. 5, the model generation unit 22 performs machine learning to generate an encoder-decoder model using the associated population transition data and type code (data D1 shown in Fig. 6(a)) as input values ​​to the encoder-decoder model and as output values ​​(correct answers) of the encoder-decoder model.

[0037] Furthermore, the model generation unit 22 may generate multiple encoder-decoder models according to the type code. For example, the model generation unit 22 may generate an encoder-decoder model for each type code. The model generation unit 22 uses data on population trends for areas and days with the same type code, as shown in FIG. 6(b), to generate one encoder-decoder model. That is, the model generation unit 22 filters the population trend data for each type code, and uses the filtered population trend data to generate an encoder-decoder model.

[0038] In this case, the model generation unit 22 may generate an encoder-decoder model (an encoder-decoder model that does not input a type code) that receives only population transition data as input, as shown in Fig. 4. The model generation unit 22 performs machine learning to generate the encoder-decoder model, using the population transition data as an input value to the encoder-decoder model and as an output value (correct answer) of the encoder-decoder model.

[0039] Alternatively, the model generation unit 22 may generate an encoder-decoder model that inputs a type code in addition to the population transition data as shown in Fig. 5. In this case, the model generation unit 22 performs machine learning to generate an encoder-decoder model, using the corresponding population transition data and type code (data D2 shown in Fig. 6(b)) as input values ​​to the encoder-decoder model and as output values ​​(correct answers) of the encoder-decoder model. The model generation unit 22 performs machine learning as described above for each type code, and generates an encoder-decoder model for each type code.

[0040] The model generation unit 22 outputs the generated encoder-decoder model to the population status determination system 10. When the model generation unit 22 generates an encoder-decoder model for each type code, it also outputs the type code corresponding to each encoder-decoder model to the population status determination system 10. These are the functions of the model generation system 20 according to this embodiment.

[0041] Next, we will explain the functions of the population status determination system 10. As shown in Figure 1, the population status determination system 10 is configured to include an acquisition unit 11, a model calculation unit 12, a determination unit 13, and a determination criterion generation unit 14.

[0042] The acquisition unit 11 is a functional unit that acquires population information indicating the time-series population of an area for which the population status is to be determined. The acquisition unit 11 acquires the above-mentioned population information for the area and time period for which the status is to be determined. The acquisition unit 11 may also acquire type information indicating the type of the area for which the population status is to be determined.

[0043] For example, the acquisition unit 11 accepts a specification of the area and time period (date) to be determined from a user of the population status determination system 10, and acquires population information relating to the specified area and time period in the same manner as the acquisition of learning population information by the learning acquisition unit 21 described above.

[0044] The acquisition unit 11 may acquire type information indicating the type of an area for which the population status is to be determined. The acquisition unit 11 acquires type information for the area related to the acquired population information that is the same as the learning type information acquired by the learning acquisition unit 21. For example, when the learning acquisition unit 21 acquires the pre-stored learning type information shown in FIG. 3(c) described above, the acquisition unit 11 acquires, from the same information, a type code corresponding to the mesh code of the area related to the population information, as type information.

[0045] In addition, when clustering is performed by the learning acquisition unit 21, the acquisition unit 11 acquires, as type information, a type code corresponding to the mesh code of the area related to the population information from the information corresponding to the mesh code and type code stored in the computer 1 as a result of the clustering.

[0046] The acquisition unit 11 outputs the acquired population information to the model calculation unit 12 and the determination unit 13. When acquiring type information, the acquisition unit 11 also outputs the acquired type information to the model calculation unit 12.

[0047] The model calculation unit 12 is a functional unit that inputs the population information acquired by the acquisition unit 11 into a pre-stored encoder-decoder model, performs calculations, and obtains output from the encoder-decoder model. The model calculation unit 12 may perform calculations using the encoder-decoder model based on the type information acquired by the acquisition unit 11. The model calculation unit 12 may also input the type information into the encoder-decoder model and obtain output from the encoder-decoder model. The model calculation unit 12 may select an encoder-decoder model to use for calculation from a plurality of pre-stored encoder-decoder models based on the type information, and perform calculations using the selected encoder-decoder model.

[0048] The model calculation unit 12 receives and stores the encoder-decoder model generated by the model generation system 20. The model calculation unit 12 receives as input the population information from the acquisition unit 11.

[0049] The model calculation unit 12 uses the population information as an input value to the encoder-decoder model and performs calculations using the weights w of the encoder-decoder model to obtain output values ​​from the encoder-decoder model. The output values ​​from the encoder-decoder model are restored data of population transition data, which is population information, and are information in the same format as the population information. Graph G2 shows an example of output values ​​when the population information shown by graph G1 in Figure 2 is used as an input value.

[0050] In an aspect in which type information is used, the model calculation unit 12 receives the type information from the acquisition unit 11 and performs the following processing. In this case, for example, an encoder-decoder model that also receives type information as input is generated by the model generation system 20 as described above. The model calculation unit 12 performs calculations using the population information and type information as input values ​​to the encoder-decoder model and the weight w of the encoder-decoder model to obtain an output value from the encoder-decoder model.

[0051] In this case, for example, as described above, the model generation system 20 generates a plurality of encoder-decoder models corresponding to the type code. The model calculation unit 12 selects, from the plurality of encoder-decoder models, an encoder-decoder model corresponding to the type code, which is the input type information. Using the selected encoder-decoder model, the model calculation unit 12 obtains an output value from the encoder-decoder model in the same manner as described above.

[0052] The model calculation unit 12 outputs the output value from the obtained encoder-decoder model to the determination unit 13. Note that the output value output to the determination unit 13 may be only the portion corresponding to the population information.

[0053] The determination unit 13 is a functional unit that compares the population information acquired by the acquisition unit 11 with the output obtained by the model calculation unit 12 to determine the population status of the area. The determination by the determination unit 13 is, for example, as described above, a determination of whether the population in the area to be determined is in an abnormal state different from normal. However, any determination other than the above may be made as long as it can be made by comparing the population information input to the encoder-decoder model with the output from the encoder-decoder model. The determination unit 13 determines the population status of the area as follows:

[0054] The determination unit 13 receives population information from the acquisition unit 11. The determination unit 13 receives an output value corresponding to the population information from the model calculation unit 12. The determination unit 13 compares the input to the encoder-decoder model (data on population transitions, for example, graph G1 in FIG. 2) with the output from the encoder-decoder model (reconstructed data on population transitions, for example, graph G2 in FIG. 2), and calculates an error as the degree of anomaly. For example, the determination unit 13 calculates the absolute value of the difference between the input and output for each time period every hour, and determines the total for all time periods as the error.

[0055] The determination unit 13 compares the calculated error with a preset threshold. If the error is equal to or greater than the threshold, the determination unit 13 determines that the population in the area being determined is in an abnormal state. In this case, it is estimated that an event or other phenomenon that differs from normal times is occurring in the area being determined. If the error is not equal to or greater than the threshold, the determination unit 13 determines that the population in the area being determined is not in an abnormal state.

[0056] The above judgment utilizes the fact that if an encoder-decoder model is generated by machine learning using only normal data, the model cannot be successfully restored when abnormal data is input to the encoder-decoder model. Therefore, under normal circumstances, the judgment is based on the learning population information used when the model generation system 20 generated the encoder-decoder model.

[0057] The determination unit 13 outputs information indicating the determination result. For example, the determination unit 13 may display the determination result on a display device provided in the computer 1 so that the user can refer to the determination result. Alternatively, the determination unit 13 may transmit the information indicating the determination result to another device. Furthermore, the determination unit 13 may output the information indicating the determination result to an output destination other than the above by a method other than the above.

[0058] The judgment criterion generation unit 14 is a functional unit that generates a judgment criterion used for judgment by the judgment unit 13. The judgment criterion generation unit 14 acquires population information for generating a judgment criterion, including first population information for generating a judgment criterion in a first state of the same area and a second state different from the first state, and second population information for generating a judgment criterion in the first state of the area to be judged, inputs the acquired first and second population information to a pre-stored encoder-decoder model for generating a judgment criterion, performs calculations, obtains an output from the encoder-decoder model, compares the input and output of the encoder-decoder model, and generates a judgment criterion based on the comparison result.

[0059] The first state may be normal, and the second state may be abnormal. The determination unit 13 may use a threshold as a determination criterion to determine whether the population is in an abnormal state different from normal, and the determination criterion generation unit 14 may generate the threshold. The determination criterion generation unit 14 may generate the threshold from a ratio of values ​​based on comparison results for the first population information in the first state and the second state, and a value based on a comparison result for the second population information. The determination criterion generation unit 14 may acquire first population information for each of a plurality of areas and generate a threshold from a statistical value of the ratio of values ​​based on comparison results for each of the plurality of areas. The determination criterion generation unit 14 may input the first population information to an encoder-decoder model for generating a determination criterion generated by machine learning from the first population information for generating a determination criterion in the first state, and input the second population information to an encoder-decoder model for generating a determination criterion generated by machine learning from the second population information for generating a determination criterion.

[0060] Since the characteristics of population transition differ from area to area, the anomaly score, which is an error calculated by the determination unit 13, also varies from area to area. Therefore, if the threshold value used in the determination by the determination unit 13 is a common value for all areas, there is a risk that an appropriate determination will not be made. In order for the determination unit 13 to make an appropriate determination, the determination criterion generation unit 14 generates a threshold value, which is a determination criterion used in the determination by the determination unit 13, for each area to be determined. For example, the determination criterion generation unit 14 generates a determination criterion as follows prior to the determination by the determination unit 13.

[0061] The criterion generation unit 14 acquires first population information and second population information, which are population information for generating the criterion. FIG. 7 schematically shows the first population information and the second population information. The training examples shown in FIG. 7 are the first population information, and the test examples are the second population information. The training examples and the test examples may be configured to include multiple individual pieces of population information in a format similar to the population information used to determine the population status.

[0062] A training example is composed of population information for multiple areas. "Case 1," "Case 2," ... "Case N-1" shown in Figure 7 represent the population information for each case. The multiple cases differ from each other in either area or time. The population information for each training example includes multiple units of population information, including population information for a first state (normal) and population information for a second state (abnormal). The unit of population information is the same as the population information used to determine the population state or generate the encoder-decoder model, for example, daily, as described above. The graph in Figure 7 shows the population information for one training example. In this graph, the horizontal axis represents time and the vertical axis represents population.

[0063] The population information during abnormal times in the training examples is, for example, the population information on the day an event occurred in the area corresponding to the population information, as shown in FIG. 7. This is because the population trend in the area on the day the event occurred is significantly different from the population trend during normal times. The population information during normal times in the training examples is, for example, the population information for multiple days (e.g., approximately 30 days) prior to the day on which the event occurred in the area (when no event occurred in the area), as shown in FIG. 7. The graph in FIG. 7 shows population information during abnormal times for one consecutive day together with population information during normal times for multiple days. It is possible to identify in advance whether the population information in the training examples relates to abnormal times or normal times. As described above, the training examples for each case include one day's population information during abnormal times and multiple days' population information during normal times. Note that the training examples for each area may include multiple days' population information during abnormal times, or may include only one day's population information during normal times.

[0064] The test case is composed of, for example, population information of an area to be judged by the judgment unit 13. Note that the test case is to judge an area related to the population information, but the population information itself is not the object of judgment. "Case N" shown in FIG. 7 is the population information of the area to be judged. The population information of the area to be judged, which is the test case, includes population information in normal times, which is the first state. The unit of the population information is the same as the population information used to judge the population state or generate the encoder-decoder model, for example, daily as described above.

[0065] The normal population information of the test case is population information for multiple days (e.g., about 30 days) when no events have occurred in the area being evaluated. In this way, the test case includes normal population information for multiple days. Note that the test case may also include normal population information for only one day.

[0066] The criterion generation unit 14 acquires training examples and test examples, which are population information for generating a criterion, in the same manner as the acquisition of training population information by the training acquisition unit 21 or the acquisition of population information by the acquisition unit 11. The criterion generation unit 14 generates a threshold value from the acquired training examples and test examples as follows.

[0067] The criterion generation unit 14 generates an encoder-decoder model for generating a criterion by machine learning from population information during normal times among the training examples. The encoder-decoder model for generating a criterion may be in the same format as the encoder-decoder model used for judgment by the judgment unit 13. The generation of the encoder-decoder model for generating a criterion may be performed in the same manner as the generation of the encoder-decoder model by the model generation unit 22 described above. The judgment criterion generation unit 14 stores the generated encoder-decoder model for generating a criterion and uses it to generate a threshold. Note that the encoder-decoder model used for judgment by the judgment unit 13 may be the encoder-decoder model for generating a criterion.

[0068] For each piece of daily population information included in the training examples, the criterion generation unit 14 uses the population information as an input value to the encoder-decoder model for generating the criterion, performs a calculation using the weights w of the encoder-decoder model for generating the criterion, and obtains an output value from the encoder-decoder model for generating the criterion. The criterion generation unit 14 compares the input to the encoder-decoder model for generating the criterion with the output from the encoder-decoder model for generating the criterion, and calculates an anomaly score, which is an error, as an anomaly score. For example, the criterion generation unit 14 calculates the absolute value of the difference between the input and output for each time period every hour, and calculates the sum of all the time periods as the anomaly score, which is an error. This error calculation is performed in the same manner as the calculation of the error used in the judgment by the judgment unit 13 described above.

[0069] The determination criterion generation unit 14 calculates, from the calculated anomaly scores, a ratio α of the maximum anomaly score in normal times (before the event) to the maximum anomaly score in abnormal times (after the event) for each case i (Case 1 to Case N-1) of the training examples, using the following formula: i Calculate.

number

[0070] In addition, the ratio α i The value used for may be a value other than the maximum value of the anomaly score for each case i as described above. For example, a preset quantile of the anomaly score for each case i may be used.

[0071] Next, the criterion generating unit 14 calculates the α i For example, when the criterion generating unit 14 uses the average value of α with a hat, it calculates α with a hat by the following formula.

number

[0072] Furthermore, the criterion generation unit 14 generates an encoder-decoder model for generating a criterion (different from the one generated from the training examples) by machine learning from population information under normal conditions, which is a test example. This encoder-decoder model for generating a criterion may also be in the same format as the encoder-decoder model used for judgment by the judgment unit 13. The generation of this encoder-decoder model for generating a criterion may also be performed in the same manner as the generation of the encoder-decoder model by the model generation unit 22 described above. The judgment criterion generation unit 14 stores the generated encoder-decoder model for generating a criterion and uses it to generate a threshold. Note that the encoder-decoder model used for judgment by the judgment unit 13 may also be the encoder-decoder model for generating a criterion.

[0073] For each piece of daily population information included in the test cases, the criterion generation unit 14 uses the population information as an input value to the encoder-decoder model for generating the criterion, performs a calculation using the weight w of the encoder-decoder model for generating the criterion, and obtains an output value from the encoder-decoder model for generating the criterion. The criterion generation unit 14 compares the input to the encoder-decoder model for generating the criterion with the output from the encoder-decoder model for generating the criterion, and calculates an anomaly score, which is an error, as an anomaly score. For example, the criterion generation unit 14 calculates the absolute value of the difference between the input and output for each time period every hour, and calculates the sum of all the time periods as the anomaly score, which is an error. This error calculation is performed in the same manner as the calculation of the error used in the judgment by the judgment unit 13 described above.

[0074] The judgment criterion generation unit 14 calculates the anomaly score R of the test case from the calculated anomaly scores. N before The maximum value of max(R N before The criterion generating unit 14 calculates the threshold value (threshold ) from the calculated value by the following formula: N Calculate.

number

[0075] The judgment criterion generating unit 14 calculates the calculated threshold N The determination unit 13 receives the threshold value from the determination criterion generation unit 14. N Enter the threshold value you entered. N The population of the area is determined as described above using the above.

[0076] As described above, by using the maximum anomaly score for each case, it is possible to make a judgment based on a threshold that uses the maximum anomaly score as a reference. In addition, by calculating the threshold from the hatched α, which is the threshold coefficient and is a statistical value of the anomaly scores for multiple training cases, and the maximum anomaly score for the test cases, it is possible to determine a threshold that is suitable for judgment in the area to be judged, taking into account the population at the time of an anomaly in each case.

[0077] The encoder-decoder model for generating the above-mentioned criterion may be for each city, ward, town, or village, prefecture, or region within a prefecture. Furthermore, the area related to the training examples ("Case 1," "Case 2," ... "Case N-1") may or may not include the area related to the test examples. For example, the areas related to the training examples may be Tokyo, Chiba, and Saitama prefectures, and the test examples may be Kanagawa prefecture, and anomaly detection may be performed in Kanagawa prefecture (when applied to different areas). Alternatively, the areas related to the training examples may be Tokyo, Chiba, and Saitama prefectures, and the test examples may be Tokyo, and anomaly detection may be performed in Tokyo (when applied to the same area). Since this embodiment aims to take into account the characteristics of each area, it is expected that appropriate thresholds can be set, especially when applied to different areas.

[0078] If the determination unit 13 makes a determination using type information such as a type code, the determination criterion generation unit 14 may also generate a threshold using type information. The type information may be used by the determination criterion generation unit 14 in the same way as by the determination unit 13. The functions of the population status determination system 10 according to this embodiment have been described above.

[0079] Next, the processing executed by the computer 1 according to this embodiment (the operating method performed by the computer 1) will be described using the flowcharts of Figures 8 to 10. First, the processing executed by the model generation system 20 will be described using the flowchart of Figure 8.

[0080] In this process, first, the learning acquisition unit 21 acquires learning population information (S01). Next, the learning acquisition unit 21 acquires learning type information (S02). Note that in a mode in which learning type information is not used, the acquisition of learning type information (S02) does not need to be performed. Next, the model generation unit 22 performs machine learning based on the learning population information to generate an encoder-decoder model (S03). Also, in a mode in which learning type information is used, an encoder-decoder model is generated based on the learning type information. The generated encoder-decoder model is output to the population state determination system 10 and stored by the model calculation unit 12. The above is the process executed by the model generation system 20 according to this embodiment.

[0081] Next, the processing executed by the population status determination system 10 will be described using the flowchart in Fig. 9. In this processing, first, the determination criterion generation unit 14 generates a threshold value, which is a determination criterion used for determination by the determination unit 13 (S11). An example of the processing of generating a threshold value, which is a determination criterion, by the determination criterion generation unit 14 (S11) will be described using the flowchart in Fig. 10.

[0082] In this process, training examples and test examples are first obtained (S111). Next, machine learning is performed using the normal population information of the training examples to generate an encoder-decoder model for generating a criterion (S112). Next, each piece of population information in the training examples is input to the encoder-decoder model for generating a criterion generated in S112, and calculations are performed to obtain an output from the encoder-decoder model (S113). Next, the input to the encoder-decoder model and the output from the encoder-decoder model are compared to calculate an anomaly score for each piece of population information in the training examples (S114).

[0083] Next, α for each example i in the training examples is calculated from the calculated anomaly score. i is calculated (S115). i From the statistical value of , a hatched α is calculated (S116).

[0084] Next, machine learning is performed using the population information of the test cases, and an encoder-decoder model for generating a criterion is generated (S117). Next, each piece of population information of the test cases is input to the encoder-decoder model for generating a criterion generated in S117, and calculations are performed, and an output from the encoder-decoder model is obtained (S118). Next, the input to the encoder-decoder model is compared with the output from the encoder-decoder model, and an anomaly score for each piece of population information of the test cases is calculated (S119). Next, a threshold value α (with a hat) calculated in S116 and the anomaly score of the test cases calculated in S119 is used to calculate the threshold value α (with a hat) N is calculated (S120). N is output from the judgment criterion generating unit 14 to the judgment unit 13, stored in the judgment unit 13, and used in the judgment by the judgment unit 13 below.

[0085] Next, as shown in Fig. 9, the acquisition unit 11 acquires population information and type information (S12). Note that in a mode where type information is not used, acquisition of type information does not have to be performed. Next, the model calculation unit 12 inputs the population information into the encoder-decoder model, performs calculations, and obtains output from the encoder-decoder model (S13). Also, in a mode where type information is used, calculations are performed using the encoder-decoder model based on the type information.

[0086] Next, the determination unit 13 compares the input to the encoder-decoder model with the output from the encoder-decoder model (S14). Next, the determination unit 13 determines the population status of the area based on the above comparison (S15). Next, the determination unit 13 outputs information indicating the determination result (S16). The above is the process executed by the population status determination system 10 according to this embodiment.

[0087] The population status determination system 10 according to this embodiment uses time-series population information, making it possible to determine the population status taking into account the time-series population of an area. Furthermore, the input to the encoder-decoder model is compared with the output to make a determination. Furthermore, appropriate determination criteria are generated based on the training examples and test examples, which are the first and second population information used to generate the determination criteria, and are used for the determination. The determination criteria thus generated take into account the characteristics of population transitions for each area, as described above. Therefore, the population status determination system 10 according to this embodiment uses appropriate determination criteria according to the area to be determined, making it possible to accurately and appropriately determine the population status.

[0088] Also, as in this embodiment, the first state may be normal and the second state may be abnormal. Furthermore, as in this embodiment, the judgment criterion may be a threshold, and the judgment of the area to be judged may be a judgment of whether the population is in an abnormal state different from normal. With this configuration, it is possible to appropriately detect abnormalities in population trends in the area to be judged, as in this embodiment. However, the first state and the second state do not necessarily have to be as described above, and may be any two states related to the judgment of the population state. Furthermore, the generated judgment criterion may be something other than a threshold, and a judgment other than the above may be made.

[0089] In addition, as mentioned above, the threshold is generated by using the ratio of the anomaly scores, α, which is the result of comparing the input and output of the encoder-decoder model for generating the judgment criteria in the training examples. i , and the anomaly score R in the training examples N before may be used. With this configuration, it is possible to generate a threshold value appropriately and reliably, and as a result, it is possible to determine the population status appropriately and reliably. Furthermore, the training examples may relate to multiple areas. This allows the hatched α used to calculate the threshold to take multiple areas into consideration, making the threshold value more appropriate. However, the threshold does not necessarily have to be generated as described above, and it may be generated using first and second population information for generating a criterion, such as training examples and test examples.

[0090] Furthermore, the encoder-decoder models for generating the criterion used to generate the criterion may be generated by machine learning from the first and second population information for generating the criterion, as described above. This configuration allows the criterion to be generated appropriately and reliably. However, the encoder-decoder models for generating the criterion do not necessarily have to be generated by machine learning from the first and second population information for generating the criterion, and may be any models that can be used to generate the criterion.

[0091] Furthermore, type information may be used as in the above-described embodiment. By using type information, it is possible to appropriately determine the population status according to the characteristics of the area. For example, it is possible to make a determination that takes into account the functional characteristics of a city, such as an office district or a residential district. This allows for more accurate and appropriate determination compared to a determination based on average population fluctuations that does not take type information into account.

[0092] The type information may be used as an input to an encoder-decoder model as described above. Alternatively, the encoder-decoder model to be used for calculation may be selected based on the type information. With such a configuration, the type information can be used reliably and appropriately, and a determination can be made reliably and appropriately. However, the type information may be used in a manner other than the above. Also, the type information does not necessarily have to be used.

[0093] The model generation system 20 according to this embodiment can generate an encoder-decoder model for use in the population status determination system 10. Furthermore, when generating the encoder-decoder model, training type information corresponding to the type information may also be used. Furthermore, the training type information may be acquired by performing clustering using the training population information, as described above. With this configuration, even if type information is not associated with an area in advance, it is possible to generate an encoder-decoder model based on the area type and perform determination using the encoder-decoder model.

[0094] In this embodiment, the computer 1 includes the population status determination system 10 and the model generation system 20, but the population status determination system 10 and the model generation system 20 may be implemented independently of each other.

[0095] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wires, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0096] Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, election, establishment, comparison, assumption, expectation, regard, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0097] For example, the computer 1 according to an embodiment of the present disclosure may function as a computer that performs information processing according to the present disclosure. Fig. 11 is a diagram illustrating an example of a hardware configuration of the computer 1 according to an embodiment of the present disclosure. The computer 1 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, etc.

[0098] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of computer 1 may be configured to include one or more of the apparatuses shown in the drawings, or may be configured to exclude some of the apparatuses.

[0099] Each function in computer 1 is realized by loading specific software (programs) onto hardware such as processor 1001 and memory 1002, causing processor 1001 to perform calculations, control communication via communication device 1004, and control at least one of reading and writing data in memory 1002 and storage 1003.

[0100] The processor 1001 controls the entire computer by running, for example, an operating system. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, each function of the computer 1 described above may be realized by the processor 1001.

[0101] Furthermore, the processor 1001 reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002, and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, each function of the computer 1 may be implemented by a control program stored in the memory 1002 and running on the processor 1001. While the above-described various processes have been described as being executed by one processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0102] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for performing information processing according to an embodiment of the present disclosure.

[0103] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray® disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy disk, a magnetic strip, etc. Storage 1003 may also be called an auxiliary storage device. The storage medium provided in computer 1 may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0104] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0105] The input device 1005 is an input device (for example, a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (for example, a touch panel).

[0106] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0107] Furthermore, the computer 1 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0108] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0109] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0110] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0111] Each aspect / embodiment described in this disclosure may be used alone, in combination, or switched depending on the implementation. Furthermore, notification of predetermined information (e.g., notification that "X is true") is not limited to being done explicitly, but may be done implicitly (e.g., by not notifying the predetermined information).

[0112] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0113] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0114] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0115] As used in this disclosure, the terms "system" and "network" are used interchangeably.

[0116] Furthermore, the information, parameters, etc. described in this disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information.

[0117] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0118] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0119] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0120] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0121] When used in this disclosure, the terms "include," "including," and variations thereof are intended to be inclusive, similar to the term "comprising." Furthermore, when used in this disclosure, the term "or" is not intended to be an exclusive or.

[0122] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0123] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0124] The population status determination system of the present disclosure has the following configuration. [1] an acquisition unit that acquires population information indicating the time series population of an area for which the population status is to be determined; a model calculation unit that inputs the population information acquired by the acquisition unit into a pre-stored encoder-decoder model that compresses and decompresses input data, performs calculations, and obtains an output from the encoder-decoder model; a determination unit that compares the population information acquired by the acquisition unit with the output obtained by the model calculation unit to determine the population status of the area; a judgment criterion generation unit that generates a judgment criterion used in the judgment by the judgment unit, The criterion generation unit acquires population information for generating a criterion, including first population information for generating a criterion in a first state of the same area and a second state different from the first state, and second population information for generating a criterion in the first state of the area to be judged, inputs the acquired first and second population information into a pre-stored encoder-decoder model for generating a criterion, performs calculations, obtains output from the encoder-decoder model, compares the input and output to the encoder-decoder model, and generates a criterion based on the comparison result. [2] The population status determination system according to [1], wherein the first status is a normal status and the second status is an abnormal status. [3] The determination unit determines whether the population is in an abnormal state different from normal using a threshold value as the determination criterion; The population status determination system according to [2], wherein the determination criterion generation unit generates the threshold value. [4] The population status determination system described in [3], wherein the determination criterion generation unit generates the threshold value from a ratio of values ​​based on the comparison results for the first population information in the first state and the second state, and a value based on the comparison results for the second population information. [5] The population status determination system described in [4], wherein the determination criterion generation unit acquires first population information for each of a plurality of areas and generates the threshold value from a statistical value of the ratio of values ​​based on the comparison results for each of the plurality of areas. [6] A population status determination system according to any one of [1] to [5], wherein the criterion generation unit inputs the first population information into an encoder-decoder model for generating a criterion generated by machine learning from the first population information for generating a criterion in the first state, and inputs the second population information into an encoder-decoder model for generating a criterion generated by machine learning from the second population information for generating a criterion. [7] The acquisition unit acquires type information indicating the type of area to be determined for the population status, The population status determination system according to any one of [1] to [6], wherein the model calculation unit performs calculations using the encoder-decoder model based on the type information acquired by the acquisition unit. [8] The population status determination system according to [7], wherein the model calculation unit also inputs the type information into the encoder-decoder model to obtain an output from the encoder-decoder model. [9] The population status determination system according to [7] or [8], wherein the model calculation unit selects an encoder-decoder model to be used for calculation from a plurality of pre-stored encoder-decoder models based on the type information, and performs calculation using the selected encoder-decoder model. [Explanation of symbols]

[0125] 1...computer, 10...population status determination system, 11...acquisition unit, 12...model calculation unit, 13...determination unit, 14...determination criterion generation unit, 20...model generation system, 21...learning acquisition unit, 22...model generation unit, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.

Claims

1. an acquisition unit that acquires population information indicating the time-series population of an area that is the target of determining the population status; a model calculation unit that inputs the population information acquired by the acquisition unit into a pre-stored encoder-decoder model that compresses and decompresses input data, performs calculations, and obtains an output from the encoder-decoder model; a determination unit that compares the population information acquired by the acquisition unit with the output obtained by the model calculation unit to determine the population status of the area; a judgment criterion generation unit that generates a judgment criterion used in the judgment by the judgment unit, The criterion generation unit acquires population information for generating a criterion, including first population information for generating a criterion in a first state of the same area and a second state different from the first state, and second population information for generating a criterion in the first state of the area to be judged, inputs the acquired first and second population information into a pre-stored encoder-decoder model for generating a criterion, performs calculations, obtains output from the encoder-decoder model, compares the input and output to the encoder-decoder model, and generates a criterion based on the comparison result. This is a population state judgment system.

2. 2. The population condition determination system according to claim 1, wherein the first condition is a normal condition, and the second condition is an abnormal condition.

3. The determination unit determines whether the population is in an abnormal state different from normal using a threshold value as the determination criterion, The population status determination system according to claim 2 , wherein the determination criterion generating unit generates the threshold value.

4. The population status determination system according to claim 3, wherein the determination criterion generation unit generates the threshold value from a ratio of values ​​based on a comparison result for the first population information in the first state and the second state, and a value based on a comparison result for the second population information.

5. The population status determination system of claim 4, wherein the determination criterion generation unit acquires first population information for each of a plurality of areas and generates the threshold value from a statistical value of a ratio of values ​​based on the comparison results for each of the plurality of areas.

6. 2. The population status determination system of claim 1, wherein the criterion generation unit inputs the first population information to an encoder-decoder model for generating a criterion generated by machine learning from first population information for generating a criterion in a first state, and inputs the second population information to an encoder-decoder model for generating a criterion generated by machine learning from second population information for generating a criterion.

7. the acquisition unit acquires type information indicating a type of area that is a target for determining a population state; The population status determination system according to claim 1 , wherein the model calculation unit performs calculations using the encoder-decoder model based on the type information acquired by the acquisition unit.

8. 8. The population status determination system according to claim 7, wherein the model calculation unit also inputs the type information into the encoder-decoder model to obtain an output from the encoder-decoder model.

9. 8. The population status determination system according to claim 7, wherein the model calculation unit selects an encoder-decoder model to be used for calculation from a plurality of pre-stored encoder-decoder models based on the type information, and performs calculation using the selected encoder-decoder model.

Images