WO2024024201A1 - 人口状態判定システム - Google Patents
人口状態判定システム Download PDFInfo
- Publication number
- WO2024024201A1 WO2024024201A1 PCT/JP2023/016969 JP2023016969W WO2024024201A1 WO 2024024201 A1 WO2024024201 A1 WO 2024024201A1 JP 2023016969 W JP2023016969 W JP 2023016969W WO 2024024201 A1 WO2024024201 A1 WO 2024024201A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- population
- encoder
- determination
- information
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G06N3/0455—Auto-encoder networks; Encoder-decoder networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Definitions
- the present invention relates to a population status determination system that determines the population status of an area.
- Patent Document 1 a technique has been proposed for estimating the population of each area and time zone using data from a mobile terminal such as a mobile phone (see, for example, Patent Document 1).
- 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.
- a population status determination system includes an acquisition unit that acquires population information indicating a 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 above into a pre-stored encoder/decoder model that compresses and restores the input data, performs calculations, and obtains an output from the encoder/decoder model, and an acquisition unit. a determination unit that compares the acquired population information with the output obtained by the model calculation unit to determine the state of the population of the area; and a determination criterion generation unit that generates a determination criterion used for determination by the determination unit.
- the determination criterion generation unit includes population information for determination criterion generation, the first population information for determination criterion generation in each of a first state in the same area and a second state different from the first state. and second population information for determination criterion generation in the first state of the area to be determined, and store the acquired first and second population information in advance for determination criterion generation. calculate the input to the encoder/decoder model, obtain the output from the encoder/decoder model, compare the input and output to the encoder/decoder model, and set the judgment criteria based on the comparison result. generate.
- the population status determination system it is possible to determine the population status in consideration of the chronological population of an area. Further, the input to the encoder/decoder model and the output are compared to make a determination. Further, an appropriate criterion is generated based on the first and second population information for generating the criterion, and used for the determination. Therefore, according to the population status determination system according to an embodiment of the present invention, the population status can be appropriately determined.
- the state of the population can be appropriately determined.
- 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. It is a graph of an example of population information and an output value from an encoder-decoder model when the population information is used as an input value.
- FIG. 3 is a diagram showing an example of information used in a computer.
- FIG. 2 is a diagram schematically showing an example of an encoder/decoder model generated and used by a computer.
- FIG. 7 is a diagram schematically showing another example of an encoder/decoder model generated and used by a computer.
- FIG. 3 is a diagram showing an example of information used in a computer.
- 3 is a diagram schematically showing a learning example and a test example, which are first and second population information for determining criteria generation, which are used to generate determination criteria.
- 3 is a flowchart showing processing executed by the model generation system according to the embodiment of the present invention. It is a flowchart which shows the processing performed by the population status determination system concerning an embodiment of the present invention. It is a flowchart which shows the process performed at the time of determination criterion generation by the population state determination system based on embodiment of this invention.
- 1 is a diagram showing 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.
- FIG. 1 shows a computer 1 that is a population status determination system 10 and a model generation system 20 according to the present embodiment.
- the population status determination system 10 is a system (apparatus) that determines (estimates) the population status of a geographical area.
- the area to be determined is, for example, a 500 m square area divided into regions. A 1/2 area mesh may be used as the area.
- administrative divisions such as municipalities or prefectures, or preset land use classifications may be used. In the following explanation, the area will be explained as a mesh. Note that the area to be determined does not need to be the above, and can be any geographical area.
- Determination by the population state determination system 10 is performed based on population information indicating the chronological population of the area to be determined. For example, population information indicating the population for each hour on a daily basis is used for this determination, as will be described later.
- the determination is, for example, a determination as to whether the population in the area to be determined is in an abnormal state different from normal times. That is, the determination is to detect an abnormality in population trends in the area to be determined.
- An abnormal state in which the population differs from normal times is, for example, a state in which population trends are excessively different from normal population trends.
- the determination by the population state determination system 10 may be not a determination of whether or not the state is abnormal, but a determination of the degree of abnormality.
- the determination by the population status determination system 10 may be other than the above as long as it determines the population status of the area.
- the determination by the population status determination system 10 is performed by performing calculations on population information using an encoder-decoder model, which is a learned model generated by machine learning.
- the encoder/decoder model is a model that compresses and decompresses input data.
- the model generation system 20 generates an encoder/decoder model used for determination by the population status determination system 10.
- a conventional computer can be used as the computer 1 that is the population status determination system 10 and model generation system 20 according to the present embodiment. Further, the computer 1 may be a computer system including multiple computers.
- the model generation system 20 includes a learning acquisition section 21 and a model generation section 22.
- 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 perform clustering using the learning population information to acquire the learning type information.
- the learning acquisition unit 21 acquires each piece of information as follows.
- the individual population information for learning is information in the same format as the population information used to determine the state of the population.
- the population information is information indicating the population of an area every hour of the day (0:00, 1:00, ..., 23:00).
- FIG. 2 shows a part of a graph G1 as an example of population information.
- the population status determination system 10 determines the population status of the area to be 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 are not necessarily as described above.
- the large amount of learning population information is used to generate the encoder/decoder model.
- the large amount of learning population information usually includes learning population information related to a plurality of areas.
- the learning acquisition unit 21 acquires, for example, the data shown in FIG. 3(a).
- the data shown in Figure 3(a) includes a mesh code (information in the "meshcode” column), information indicating time (information in the "timestamp” column), and information indicating population (information in the "population” column). This is associated information.
- the mesh code is information such as a character string that specifies a mesh that is an area, and is set in advance for each area.
- the information indicating the time is, for example, information indicating the year, month, day, and time of the day.
- the information indicating the population indicates the population at the area and time indicated by the corresponding mesh code and information indicating the time.
- the data related to the population shown in FIG. 3(a) is generated as spatial statistical information using, for example, an existing method from information indicating the location of the mobile phone and information registered about the subscriber of the mobile phone. Moreover, the data related to the population shown in FIG. 3(a) may be generated by any method other than the above.
- the learning acquisition unit 21 acquires data related to the population shown in FIG. 3A that is stored in advance in a database of the computer 1 or other device.
- the learning acquisition unit 21 converts the acquired data into data for each mesh code and for each hour of the day (0:00, 1:00, ..., 23:00), that is, area It is formatted into data of daily population trends in units. This data on population trends corresponds to learning population information.
- the learning acquisition unit 21 acquires enough data on population trends to generate an encoder/decoder model by machine learning.
- the data on population trends may or may not include data on the area whose population status is to be determined. Note that the learning acquisition unit 21 may acquire information indicating a time-series population other than the above as the learning population information.
- the learning acquisition unit 21 may be configured to acquire learning type information indicating the type of area related to population change data.
- the type of area is a type that can affect population trends in the area.
- the area types are city types such as "office area” and "residential area.”
- the learning acquisition unit 21 acquires learning type information that is stored in advance 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) includes a mesh code (information in the "meshcode” column), information indicating the city type (information in the "city type” column), and type code (information in the "type code” column). is the associated information.
- the information indicating the city type is information indicating the meaning of the area type indicated by the corresponding mesh code.
- Information indicating the city type is set in advance for each area. Note that the information indicating the city type does not need to be used for processing in the model generation system 20, and therefore does not need to be acquired.
- the type code is information (a flag indicating the area) that specifies the type of area indicated by the corresponding mesh code, 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 if the city type is the same, and is a different numerical value if the city type is different.
- the learning acquisition unit 21 acquires, as learning type information, the type code corresponding to the mesh code of the area related to the population change data.
- the learning acquisition unit 21 may perform clustering using the learning population information to acquire the learning type information.
- the learning acquisition unit 21 performs clustering using the data on daily population trends in area units. For example, as described below, the learning acquisition unit 21 performs area clustering using the data on daily population trends in area units. By performing such clustering, areas with similar population trends can be divided into clusters.
- the learning acquisition unit 21 averages the population at each time for each area and generates one population trend data for one area. For example, the learning acquisition unit 21 averages data on daily population trends for each time during a preset period for each area (for example, from one month before the current time to the current time), and One population trend data is generated for each period.
- the learning acquisition unit 21 performs area clustering by clustering the population change data. Clustering itself may be performed by a conventional method (eg, k-means method).
- the learning acquisition unit 21 may cluster population change data that may include a plurality of population change data for one area. The learning acquisition unit 21 determines, for each area, the cluster that includes the most population change data as the cluster for the area.
- the learning acquisition unit 21 assigns a different type code (cluster number) to each cluster.
- the learning acquisition unit 21 sets the type code of the cluster to which the area belongs as the learning type information regarding the area.
- the learning acquisition unit 21 stores the correspondence between the mesh code and the type code for each area in the computer 1 so that it can also be used by the population status determination system 10. Note that when learning type information is obtained by performing clustering, there is no information indicating the city type.
- the learning acquisition unit 21 outputs the acquired learning population information to the model generation unit 22. Furthermore, in the mode of acquiring learning type information, the learning acquisition unit 21 also outputs the acquired learning type information to the model generation unit 22 .
- 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 receives type information indicating the type of area.
- the model generation unit 22 may generate a plurality of encoder/decoder models according to type information indicating the type of area.
- FIG. 4 shows an example of an encoder/decoder model.
- the encoder/decoder model is composed of a neural network, which is trained to input population information indicating the time-series population of an area, perform dimension compression, and then output the original population information. It is a completed model.
- autoencoder GPU Hinton and Salakhutdinov Ruslan, “Reducing the dimensionality of data Science, pp. 504-507, 2006
- Transformer As an encoder/decoder model, autoencoder (Geoffrey Hinton and Salakhutdinov Ruslan, “Reducing the dimensionality of data Science, pp. 504-507, 2006), or Transformer (Ashish Vaswani et al., “Attention Is All You Need.”Advances in neural information processing system 2017) can be used.
- the input layer of the encoder-decoder model is provided with neurons as many as the number of elements of population information (the number of dimensions of population information). If the population information is information (numerical value) that indicates the population of an area every hour of the day (0:00, 1:00, ..., 23:00), the input layer of the encoder/decoder model contains information about the population for each hour. There are 24 neurons (vectors) that input the numerical value of the population of the area.
- the output layer of the encoder-decoder model is provided with the same number of neurons (vectors) as neurons in the input layer, corresponding to each neuron in the input layer.
- the configuration of the encoder/decoder model itself may be similar to a conventional encoder/decoder model.
- a hidden layer including a plurality of 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 to each other with a weight w used for calculation set.
- each neuron in the hidden layer and each neuron in the output are connected to each other with a weight w used for calculation set.
- the number of neurons provided in the hidden layer is smaller than the number of neurons in the input layer and the output layer. As a result, dimension compression is performed in the hidden layer.
- the model generation unit 22 generates an encoder/decoder model as follows. First, an example of a mode in which the learning type information is not used will be explained, and then an example of a mode in which the learning type information is used will be explained.
- the model generation unit 22 receives data on population trends, which is population information for learning, from the learning acquisition unit 21. As shown in FIG. 4, the model generation unit 22 performs machine learning to generate an encoder-decoder model using data on population trends as input values to the encoder-decoder model and output values (correct answers) of the encoder-decoder model. do.
- the above machine learning itself for generating the encoder/decoder model can be performed in the same manner as conventional machine learning methods. The above is an example where learning type information is not used.
- the model generation unit 22 receives a type code as learning type information from the learning acquisition unit 21 along with population change data. In this case, the model generation unit 22 generates an encoder/decoder model that also inputs the type code.
- FIG. 5 shows an example of this encoder/decoder model.
- this encoder/decoder model includes neurons corresponding to type codes in the input layer and the output layer.
- the model generation unit 22 associates the type code of the area with data on the area and daily population trends, as shown in FIG. 6(a). This association is performed using the mesh code as a key. As shown in FIG. 5, the model generation unit 22 inputs the associated population change data and type code (data D1 shown in FIG. 6(a)) to the encoder/decoder model, and As the output value (correct answer), machine learning is performed to generate an encoder/decoder model.
- the model generation unit 22 may generate a plurality of 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 areas with the same type code and daily population trends as shown in FIG. 6(b) to generate one encoder-decoder model. That is, the model generation unit 22 filters the population change data for each type code, and uses the filtered population change data to generate an encoder/decoder model.
- the model generation unit 22 may generate an encoder/decoder model (an encoder/decoder model that does not input a type code) that inputs only population change data as shown in FIG. 4.
- the model generation unit 22 performs machine learning to generate an encoder-decoder model by using the population change data as an input value to the encoder-decoder model and as an output value (correct answer) of the encoder-decoder model.
- the model generation unit 22 may generate an encoder/decoder model that also inputs a type code in addition to the population change data as shown in FIG.
- the model generation unit 22 inputs the data on population trends and the type code (data D2 shown in FIG. 6(b)) that are associated with each other to the encoder/decoder model, and As the output value (correct answer), machine learning is performed to generate an encoder/decoder model.
- the model generation unit 22 performs machine learning as described above for each type code to generate an encoder/decoder model for each type code.
- the model generation unit 22 outputs the generated encoder/decoder model to the population status determination system 10.
- the model generation unit 22 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 state determination system 10.
- the above are the functions of the model generation system 20 according to this embodiment.
- the population state determination system 10 includes an acquisition section 11, a model calculation section 12, a determination section 13, and a determination criterion generation section 14.
- the acquisition unit 11 is a functional unit that acquires population information indicating the time-series population of an area whose population status is to be determined.
- the acquisition unit 11 acquires the above-mentioned population information of the area and time zone to be determined.
- the acquisition unit 11 may acquire type information indicating the type of area whose population status is to be determined.
- the acquisition unit 11 receives the designation of the area and time period (date) to be determined from the user of the population status determination system 10, and acquires the population information related to the designated area and time period for the above-mentioned learning purpose.
- the information is acquired in the same manner as the learning population information is acquired by the acquisition unit 21.
- the acquisition unit 11 may acquire type information indicating the type of area whose population status is to be determined.
- the acquisition unit 11 acquires the same type information as the learning type information acquired by the learning acquisition unit 21 for the area related to the population information to be acquired. For example, when the learning acquisition unit 21 acquires the pre-stored learning type information shown in FIG. Obtain the type code corresponding to as type information.
- the acquisition unit 11 acquires areas related to population information from information on the correspondence between mesh codes and type codes stored in the computer 1 as a result of the clustering. Obtain the type code corresponding to the mesh code as type information.
- the acquisition unit 11 outputs the acquired population information to the model calculation unit 12 and determination unit 13. Further, when acquiring type information, the acquiring unit 11 also outputs the acquired type information to the model calculation unit 12.
- 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 an output from the encoder/decoder model.
- the model calculation unit 12 may perform calculation using an encoder/decoder model based on the type information acquired by the acquisition unit 11.
- the model calculation unit 12 may also input type information to the encoder/decoder model and obtain an output from the encoder/decoder model.
- the model calculation unit 12 may select an encoder/decoder model to be used for calculation from a plurality of encoder/decoder models stored in advance based on the type information, and perform calculation using the selected encoder/decoder model.
- 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 population information from the acquisition unit 11 .
- the model calculation unit 12 uses the population information as an input value to the encoder-decoder model, performs calculation using the weight w of the encoder-decoder model, and obtains an output value from the encoder-decoder model.
- the output value from the encoder-decoder model is reconstructed data of population change data, which is population information, and is information in the same format as population information.
- a graph G2 is shown as an example of an output value when the population information shown by the graph G1 shown in FIG. 2 is used as an input value.
- the model calculation unit 12 receives the type information from the acquisition unit 11 and performs the following processing. In this case, for example, as described above, the model generation system 20 generates an encoder/decoder model that also inputs type information.
- the model calculation unit 12 uses population information and type information as input values to the encoder-decoder model, performs calculations using the weight w of the encoder-decoder model, and obtains an output value from the encoder-decoder model.
- the model generation system 20 generates a plurality of encoder/decoder models corresponding to the type code.
- the model calculation unit 12 selects an encoder/decoder model corresponding to the type code that is the input type information from among the plurality of encoder/decoder models.
- the model calculation unit 12 uses the selected encoder/decoder model to obtain an output value from the encoder/decoder model in the same manner as described above.
- 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 outputted to the determination unit 13 may be only the portion corresponding to the population information.
- the determination unit 13 is a functional unit that compares the population information acquired by the acquisition unit 11 and the output obtained by the model calculation unit 12 to determine the state of the population of the area.
- the determination by the determination unit 13 is, for example, determination as to whether the population in the area to be determined is in an abnormal state different from normal times, as described above. However, other determinations may be made as long as the determination can be made by comparing the population information input to the encoder/decoder model and the output from the encoder/decoder model.
- the determination unit 13 determines the population status of the area as follows.
- the determination unit 13 receives population information from the acquisition unit 11.
- the determination unit 13 receives an output value corresponding to the above population information from the model calculation unit 12.
- the determination unit 13 receives input to the encoder/decoder model (population transition data, for example, graph G1 in FIG. 2) and output from the encoder/decoder model (restored data of population transition data, for example, the graph in FIG. 2). G2) and calculate the error as the degree of abnormality.
- the determination unit 13 calculates the absolute value of the difference between the input and output in each time period for each hour, and uses the sum of all time periods as the error.
- the determination unit 13 compares the calculated error with a preset threshold. If the error is greater than or equal to the threshold, the determining unit 13 determines that the population in the area to be determined is in an abnormal state. In this case, it is presumed that a phenomenon different from normal times, such as an event, is occurring in the area to be determined. If the error is not equal to or greater than the threshold, the determining unit 13 determines that the population in the area to be determined is not in an abnormal state.
- the above determination utilizes the fact that when an encoder/decoder model is generated by machine learning using only normal data, it cannot be successfully restored if abnormal data is input to the encoder/decoder model. Therefore, during normal times, the determination is based on the learning population information used when the model generation system 20 generates the encoder/decoder model.
- the determination unit 13 outputs information indicating the determination result.
- the determination unit 13 may display the determination result on a display device included in the computer 1 so that the user can refer to the determination result.
- the determination unit 13 may transmit information indicating the determination result to another device.
- the determination unit 13 may output information indicating the determination result to an output destination other than the above using a method other than the above.
- the determination criterion generation unit 14 is a functional unit that generates determination criteria used for determination by the determination unit 13.
- the determination criterion generation unit 14 generates first population information for determination criterion generation in a first state in the same area and in a second state different from the first state. , as well as second population information for determination criterion generation in the first state of the area to be determined, and transmit the acquired first and second population information to an encoder for determination criterion generation stored in advance.
- An input is input to a decoder model, a calculation is performed, an output is obtained from the encoder/decoder model, the input and output to the encoder/decoder model are compared, and a determination criterion is generated based on the comparison result.
- the first state may be a normal state
- the second state may be an abnormal state.
- the determination unit 13 may use a threshold value as a determination criterion to determine whether the population is in an abnormal state different from normal times, and the determination criterion generation unit 14 may generate the threshold value.
- the determination criterion generation unit 14 generates a threshold value from the ratio of values based on the comparison results in the first state and the second state for the first population information, and the value based on the comparison results for the second population information. You may.
- the determination criterion generation unit 14 may acquire the first population information of each of the plurality of areas, and generate the threshold value from the statistical value of the ratio of values based on the comparison results of each of the plurality of areas.
- the determination criterion generation unit 14 inputs the first population information into an encoder/decoder model for determination criterion generation generated by machine learning from the first population information for determination criterion generation in the first state,
- the second population information may be input into an encoder/decoder model for determination criterion generation that is generated by machine learning from the second population information for determination criterion generation.
- 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 generating section 14 generates the determination criterion as follows prior to the determination by the determining section 13.
- the determination criterion generation unit 14 acquires first population information and second population information that are population information for determination criterion generation.
- FIG. 7 schematically shows the first population information and the second population information.
- the learning example shown in FIG. 7 is the first population information
- the test example is the second population information.
- the learning example and the test example may include a plurality of pieces of individual population information in the same format as the population information used to determine the state of the population.
- the learning example is composed of, for example, population information of multiple areas.
- “Case 1”, “Case 2”, . . . “Case N-1” shown in FIG. 7 are population information for each case.
- the plurality of cases are different from each other in either area or time.
- the population information for each case which is a learning example, includes population information in a plurality of units, including population information in a first state in normal times and population information in an abnormal time in a second state.
- the unit of the population information is the same as the population information used for determining the state of the population or generating the encoder/decoder model, and is, for example, in units of one day as described above.
- the graph shown in FIG. 7 shows the population information of one case, which is a learning case. In this graph, the horizontal axis is time and the vertical axis is population.
- the population information at the time of abnormality among the learning examples is, for example, the population information on the day when an event occurred in the area corresponding to the population information, as shown in FIG. This is because the population trends in the area on the day the event occurred are excessively different from the population trends in normal times.
- the population information during normal times in the learning examples is based on multiple days (for example, about 30 days) before the day when an event occurred in the area (no event occurred in the area). ) population information.
- the graph in FIG. 7 collectively shows population information during abnormal times on one consecutive day and population information during normal times on multiple days. It is possible to identify in advance whether the population information of the learning case is related to abnormal times or normal times.
- the learning example for each case includes one day's worth of population information during abnormal times and multiple days' worth of population information during normal times.
- the learning examples for each area may include population information during abnormal times for multiple days, or may include population information during normal times for only one day.
- the test example is composed of, for example, population information of the area to be determined by the determination unit 13.
- the area related to population information is the target of determination, but the population information itself is not the target of determination.
- "Case N" shown in FIG. 7 is population information of the area to be determined.
- the population information of the area to be determined which is a test example, includes population information in a normal state, which is the first state.
- the unit of the population information is the same as the population information used for determining the state of the population or generating the encoder/decoder model, and is, for example, in units of one day as described above.
- the normal population information in the test case is population information for multiple days (for example, about 30 days) when no events occur in the area to be determined. In this way, the test case includes population information during normal times for multiple days. Note that the test example may include only one day's worth of normal population information.
- the judgment criterion generation unit 14 receives learning examples and test cases, which are population information for judgment criterion generation, in the same manner as the above-described acquisition of learning population information by the learning acquisition unit 21 or acquisition of population information by the acquisition unit 11. get.
- the criterion generation unit 14 generates a threshold value from the acquired learning examples and test examples as follows.
- the judgment criterion generation unit 14 generates an encoder/decoder model for judgment criterion generation by machine learning from normal population information among the learning examples.
- the encoder/decoder model for generating the determination criteria may be in the same format as the encoder/decoder model used for determination by the determination unit 13.
- the generation of the encoder/decoder model for generating the determination criteria may be performed in the same manner as the generation of the encoder/decoder model by the model generation unit 22 described above.
- the determination criterion generation unit 14 stores the generated encoder/decoder model for determination criterion generation, and uses it to generate a threshold value.
- the encoder/decoder model used for the determination by the determination unit 13 may be an encoder/decoder model for generating determination criteria.
- the judgment criterion generation unit 14 inputs the population information for each day of population information included in the learning example into the encoder/decoder model for judgment criterion generation as an input value to the encoder/decoder model for judgment criterion generation. A calculation using the weight w is performed to obtain an output value from the encoder/decoder model for generating the criterion. The judgment criterion generation unit 14 compares the input to the encoder/decoder model for judgment criterion generation with the output from the encoder/decoder model for judgment criterion generation, and calculates an abnormality score which is an error as the abnormality degree. do.
- the criterion generation unit 14 calculates the absolute value of the difference between the input and output for each hourly time period, and sets the sum of all time periods as the abnormality score, which is the error. This calculation of the error is performed in the same manner as the calculation of the error used in the determination by the determination unit 13 described above.
- the judgment criterion generation unit 14 calculates the maximum value of the abnormality score during normal times (before the event) and the maximum value during abnormality using the following formula for each case i (Case 1 to Case N-1) of the learning examples from the calculated abnormality score.
- the ratio ⁇ i of the maximum abnormality score (after the event) is calculated.
- R i before is each abnormality score on a daily basis during normal times (before the event) of case i
- max(R i before ) is the maximum value thereof for each case i.
- R i after is each abnormality score on a daily basis at the time of abnormality (after the event) of case i
- max(R i after ) is their maximum value for each case i.
- the ratio ⁇ i is the ratio of the maximum values based on the comparison results of the input and output of the encoder/decoder model for generating the criterion in normal times and abnormal times for the learning example.
- a value other than the maximum value of the abnormality score for each case i may be used as described above.
- a preset quantile of the abnormality score for each case i may be used.
- the criterion generation unit 14 calculates the shaded ⁇ from the statistical values of ⁇ i of the plurality of cases i. For example, in the case where the token ⁇ is an average value, the determination criterion generation unit 14 calculates the bracketd ⁇ using the following formula. Note that the bracketd ⁇ may be a statistical value of ⁇ i of a plurality of cases other than the average value, and may be, for example, a median value, a preset quantile, or the like.
- the judgment criterion generation unit 14 generates an encoder/decoder model for judgment criterion generation (different from the one generated from the learning example) by machine learning from the normal population information that is the test case.
- the encoder/decoder model for generating the determination criteria may also have the same format as the encoder/decoder model used for determination by the determination unit 13.
- the generation of the encoder/decoder model for generating the determination criteria may be performed in the same manner as the generation of the encoder/decoder model by the model generation unit 22 described above.
- the determination criterion generation unit 14 stores the generated encoder/decoder model for determination criterion generation, and uses it to generate a threshold value. Note that the encoder/decoder model used for the determination by the determination unit 13 may be the encoder/decoder model for generating this determination criterion.
- the judgment criterion generation unit 14 inputs the population information for each day of population information included in the test case into the encoder/decoder model for judgment criterion generation as an input value to the encoder/decoder model for judgment criterion generation. A calculation using the weight w is performed to obtain an output value from the encoder/decoder model for generating the criterion. The judgment criterion generation unit 14 compares the input to the encoder/decoder model for judgment criterion generation with the output from the encoder/decoder model for judgment criterion generation, and calculates an abnormality score which is an error as the abnormality degree. do.
- the criterion generation unit 14 calculates the absolute value of the difference between the input and output for each hourly time period, and sets the sum of all time periods as the abnormality score, which is the error. This calculation of the error is performed in the same manner as the calculation of the error used in the determination by the determination unit 13 described above.
- the criterion generation unit 14 calculates the maximum value max (R N before ) of the abnormality score R N before of the test case from the calculated abnormality score.
- the determination criterion generation unit 14 calculates a threshold value threshold N from the above calculated value using the following formula.
- the determination criterion generation unit 14 outputs the calculated threshold value threshold N to the determination unit 13.
- the determination unit 13 receives the threshold value threshold N from the determination criterion generation unit 14, and uses the input threshold value threshold N to determine the population of the area as described above.
- the maximum value of the abnormality score of each case it is possible to perform a determination based on a threshold value based on the case where the abnormality degree score is the maximum value.
- the threshold from the shaded ⁇ , which is the statistical value of the anomaly score of multiple learning examples, which is the threshold coefficient, and the maximum value of the anomaly score of the test cases, we can calculate the A threshold value suitable for determination in the determination target area can be set in consideration of the population.
- the encoder/decoder model for generating the above-mentioned criteria may be for each city, ward, town, or village, or for each prefecture, or for each region within a prefecture.
- the areas related to the learning cases (“Case 1”, “Case 2”, . . . “Case N-1”) may or may not include the areas related to the test cases.
- the area related to the learning example may be set as the area of Tokyo, Chiba prefecture, and Saitama prefecture
- the test case may be set as the area of Kanagawa prefecture, and anomaly detection may be performed in the area of Kanagawa prefecture (if applied to another area) ).
- the area related to the learning example may be the area of Tokyo, Chiba prefecture, and Saitama prefecture
- the test example may be the area of Tokyo
- anomaly detection may be performed in the area of Tokyo (if applied to the same area). . Since this embodiment aims to consider the characteristics of an area, it is expected that an appropriate threshold value can be set especially when applied to another area.
- the threshold value may also be generated by the determination criterion generation unit 14 using the type information.
- the method of using the type information by the determination criterion generation section 14 may be the same as that used by the determination section 13. The above are the functions of the population status determination system 10 according to this embodiment.
- learning population information is acquired by the learning acquisition unit 21 (S01).
- learning type information is acquired by the learning acquisition unit 21 (S02).
- acquisition of learning type information (S02) may not be performed.
- the model generation unit 22 performs machine learning based on the learning population information to generate an encoder/decoder model (S03).
- an encoder/decoder model is generated based on the learning type information.
- the generated encoder/decoder model is output to the population status 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.
- the criterion generating section 14 generates a threshold value that is a criterion used in the determination by the determining section 13 (S11).
- S11 a threshold value used in the determination by the determining section 13
- An example of the process of generating a threshold value (S11), which is a determination criterion, by the determination criterion generation unit 14 will be described using the flowchart of FIG. 10.
- a learning example and a test example are acquired (S111). Subsequently, machine learning is performed using the normal population information of the learning case, and an encoder/decoder model for generating a criterion is generated (S112). Subsequently, each piece of population information of the learning example is input to the encoder/decoder model for determination criterion generation generated in S112, calculation is performed, and an output from the encoder/decoder model is obtained (S113). Subsequently, the input to the encoder/decoder model is compared with the output from the encoder/decoder model, and an abnormality score for each population information of the learning example is calculated (S114).
- ⁇ i for each example i of the learning examples is calculated from the calculated abnormality score (S115). Subsequently, substituted ⁇ is calculated from the statistical values of ⁇ i of the plurality of cases i (S116).
- population information and type information are acquired by the acquisition unit 11 (S12). Note that in an embodiment in which type information is not used, type information does not need to be acquired.
- the model calculation unit 12 inputs the population information to the encoder/decoder model, performs calculations, and obtains an output from the encoder/decoder model (S13). Furthermore, in an embodiment where type information is used, calculations using an encoder/decoder model are performed based on the type information.
- the determination unit 13 compares the input to the encoder/decoder model and the output from the encoder/decoder model (S14). Subsequently, the determination unit 13 determines the population status of the area based on the above comparison (S15). Subsequently, 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.
- the population state determination system 10 since time-series population information is used, it is possible to determine the population state in consideration of the time-series population of an area. Further, the input to the encoder/decoder model and the output are compared to make a determination. Furthermore, appropriate criteria are generated based on the learning examples and test examples, which are the first and second population information for generating criteria, and used for the determination. The determination criteria generated in this manner take into consideration the characteristics of population trends in each area, as described above. Therefore, according to the population status determination system 10 according to the present embodiment, the population status can be appropriately determined with high accuracy by using appropriate criteria according to the area to be determined.
- the first state may be a normal state
- the second state may be an abnormal state
- the determination criterion is a threshold value
- the determination of the area to be determined may be a determination of whether the population is in an abnormal state different from normal times. According to this configuration, it is possible to appropriately detect abnormalities in population trends in the area to be determined, as in the present embodiment.
- the first state and the second state do not necessarily need to be as described above, and may be any two states related to the determination of the state of the population.
- the generated determination criterion may be other than the threshold value, and determinations other than those described above may be performed.
- the value ⁇ i of the ratio of the abnormality score which is the comparison result of the input and output of the encoder/decoder model for judgment criterion generation in the learning example, and An anomaly score R N before may be used.
- the threshold value can be appropriately and reliably generated, and as a result, the state of the population can be appropriately and reliably determined.
- the learning examples may relate to multiple areas.
- the shaded ⁇ used for calculating the threshold value can be set in consideration of a plurality of areas, and the threshold value can be made more appropriate.
- the generation of the threshold value does not necessarily need to be performed as described above, and may be performed using the first and second population information for generating the criterion, such as learning examples and test examples.
- the encoder/decoder models for generating the criterion used for generating the criterion may be generated by machine learning from the first and second population information for generating the criterion, as described above. According to this configuration, the determination criteria can be generated appropriately and reliably.
- the encoder/decoder models for generating judgment criteria do not necessarily have to be generated by machine learning from the first and second population information for generating judgment criteria, and can be used to generate judgment criteria. It can be anything as long as it is possible.
- type information may be used as in the embodiment described above.
- the type information it is possible to appropriately determine the state of the population 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 area or a residential area. This makes it possible to perform more accurate and appropriate determinations than determinations based on average population changes that do not take type information into consideration.
- the type information may be used as input to the encoder/decoder model as described above.
- the encoder/decoder model used for calculation may be selected based on the type information. According to such a configuration, type information can be used reliably and appropriately, and determination can be performed reliably and appropriately.
- the type information may be used in a method other than the above. Further, type information does not necessarily need to be used.
- an encoder/decoder model used in the population status determination system 10 can be generated. Also, when generating an encoder/decoder model, learning type information corresponding to the type information may be used. Further, the learning type information may be obtained by performing clustering using the learning population information as described above. According to 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 to make a determination using the encoder/decoder model.
- 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. may be done.
- each functional block may be realized using one physically or logically coupled device, or may be realized using two or more physically or logically separated devices directly or indirectly (e.g. , wired, wireless, etc.) and may be realized using a plurality of these devices.
- the functional block may be realized by combining software with the one device or the plurality of devices.
- Functions include judgment, decision, judgment, calculation, calculation, processing, derivation, investigation, exploration, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, These include, but are not limited to, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assigning. I can't.
- a functional block (configuration unit) that performs transmission is called a transmitting unit or transmitter. In either case, as described above, the implementation method is not particularly limited.
- the computer 1 in an embodiment of the present disclosure may function as a computer that performs information processing of the present disclosure.
- FIG. 11 is a diagram illustrating an example of the 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, and the like.
- the word “apparatus” can be read as a circuit, a device, a unit, etc.
- the hardware configuration of the computer 1 may be configured to include one or more of each device shown in the figure, or may be configured not to include some of the devices.
- Each function in the computer 1 is performed by loading predetermined software (programs) onto hardware such as the processor 1001 and memory 1002, so that the processor 1001 performs calculations, controls communication by the communication device 1004, and controls communication between the memory 1002 and the memory 1002. This is realized by controlling at least one of reading and writing data in the storage 1003.
- the processor 1001 for example, operates an operating system to control the entire computer.
- the processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, registers, and the like.
- CPU central processing unit
- each function in the computer 1 described above may be realized by the processor 1001.
- the processor 1001 reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 to the memory 1002, and executes various processes in accordance with these.
- programs program codes
- software modules software modules
- data etc.
- the program a program that causes a computer to execute at least part of the operations described in the above embodiments is used.
- each function in the computer 1 may be realized by a control program stored in the memory 1002 and operated in the processor 1001.
- Processor 1001 may be implemented by one or more chips. Note that the program may be transmitted from a network via a telecommunications line.
- the memory 1002 is a computer-readable recording medium, and includes at least one of ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), and RAM (Random Access Memory). may be done.
- Memory 1002 may be called a register, cache, main memory, or the like.
- the memory 1002 can store executable programs (program codes), software modules, and the like to implement information processing according to an embodiment of the present disclosure.
- the storage 1003 is a computer-readable recording medium, such as an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, or a magneto-optical disk (for example, a compact disk, a digital versatile disk, or a Blu-ray disk). (registered trademark disk), smart card, flash memory (eg, card, stick, key drive), floppy disk, magnetic strip, etc.
- Storage 1003 may also be called an auxiliary storage device.
- the storage medium included in the computer 1 may be, for example, a database including at least one of the memory 1002 and the storage 1003, a server, or other appropriate medium.
- 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 referred to as a network device, network controller, network card, communication module, etc., for example.
- the input device 1005 is an input device (eg, keyboard, mouse, microphone, switch, button, sensor, etc.) that accepts input from the outside.
- the output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that performs output to the outside. Note that the input device 1005 and the output device 1006 may have an integrated configuration (for example, a touch panel).
- 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 for each device.
- the computer 1 also includes hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA).
- DSP digital signal processor
- ASIC application specific integrated circuit
- PLD programmable logic device
- FPGA field programmable gate array
- a part or all of each functional block may be realized by the hardware.
- processor 1001 may be implemented using at least one of these hardwares.
- the input/output information may be stored in a specific location (for example, memory) or may be managed using a management table. Information etc. to be input/output may be overwritten, updated, or additionally written. The output information etc. may be deleted. The input information etc. may be transmitted to other devices.
- Judgment may be made using a value expressed by 1 bit (0 or 1), a truth value (Boolean: true or false), or a comparison of numerical values (for example, a predetermined value). (comparison with a value).
- notification of prescribed information is not limited to being done explicitly, but may also be done implicitly (for example, not notifying the prescribed information). Good too.
- Software includes instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, whether referred to as software, firmware, middleware, microcode, hardware description language, or by any other name. , should be broadly construed to mean an application, software application, software package, routine, subroutine, object, executable, thread of execution, procedure, function, etc.
- software, instructions, information, etc. may be sent and received via a transmission medium.
- a transmission medium For example, if the software uses wired technology (coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), etc.) and/or wireless technology (infrared, microwave, etc.) to create a website, When transmitted from a server or other remote source, these wired and/or wireless technologies are included within the definition of transmission medium.
- wired technology coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), etc.
- wireless technology infrared, microwave, etc.
- system and “network” are used interchangeably.
- information, parameters, etc. described in this disclosure may be expressed using absolute values, relative values from a predetermined value, or using other corresponding information. may be expressed.
- determining may encompass a wide variety of operations.
- “Judgment” and “decision” include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, search, and inquiry. (e.g., searching in a table, database, or other data structure), and regarding an ascertaining as a “judgment” or “decision.”
- judgment and “decision” refer to receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, and access.
- (accessing) may include considering something as a “judgment” or “decision.”
- judgment and “decision” refer to resolving, selecting, choosing, establishing, comparing, etc. as “judgment” and “decision”. may be included.
- judgment and “decision” may include regarding some action as having been “judged” or “determined.”
- judgment (decision) may be read as “assuming", “expecting", “considering”, etc.
- connection refers to any connection or coupling, direct or indirect, between two or more elements and to each other. It may include the presence of one or more intermediate elements between two elements that are “connected” or “coupled.”
- the bonds or connections between elements may be physical, logical, or a combination thereof. For example, "connection” may be replaced with "access.”
- two elements may include one or more electrical wires, cables, and/or printed electrical connections, as well as in the radio frequency domain, as some non-limiting and non-inclusive examples. , electromagnetic energy having wavelengths in the microwave and optical (both visible and non-visible) ranges.
- the phrase “based on” does not mean “based solely on” unless explicitly stated otherwise. In other words, the phrase “based on” means both “based only on” and “based at least on.”
- any reference to elements using the designations "first,” “second,” etc. does not generally limit the amount or order of those elements. These designations may be used in this disclosure as a convenient way to distinguish between two or more elements. Thus, reference to a first and second element does not imply that only two elements may be employed or that the first element must precede the second element in any way.
- a and B are different may mean “A and B are different from each other.” Note that the term may also mean that "A and B are each different from C”. Terms such as “separate” and “coupled” may also be interpreted similarly to “different.”
- the population status determination system of the present disclosure has the following configuration.
- An acquisition unit that acquires population information indicating a 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 restores 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 and the output obtained by the model calculation unit to determine the state of the population of the area; a determination criterion generation unit that generates a determination criterion used in the determination by the determination unit, The determination criterion generation unit generates population information for determination criterion generation, the first population information for determination criterion generation in each of a first state in the same area and a second state different from the first state.
- the determination unit determines whether the population is in an abnormal state different from normal times, 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.
- the determination criterion generation unit generates a ratio of values based on the comparison results in the first state and the second state for the first population information, and a value based on the comparison results for the second population information.
- the population status determination system according to [3], which generates the threshold value from.
- the criterion generation unit obtains first population information for each of the plurality of areas, and generates the threshold from a statistical value of a ratio of values based on the comparison results for each of the plurality of areas.
- the judgment criterion generation unit inputs the first population information to an encoder/decoder model for judgment criterion generation, which is generated by machine learning from the first population information for judgment criterion generation in the first state.
- input the second population information to the encoder/decoder model for judgment criterion generation generated by machine learning from the second population information for judgment criterion generation. Described population status determination system.
- the acquisition unit acquires type information indicating the type of area whose population status is to be determined;
- the population status determination system according to any one of [1] to [6], wherein the model calculation unit performs calculation using the encoder/decoder model based on the type information acquired by the acquisition unit.
- 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.
- the population status determination system according to [7] or [8].
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
[1]人口の状態の判定対象となるエリアの時系列の人口を示す人口情報を取得する取得部と、
前記取得部によって取得された人口情報を、入力データを圧縮して復元する予め記憶したエンコーダ・デコーダモデルに入力して演算を行って、当該エンコーダ・デコーダモデルからの出力を得るモデル演算部と、
前記取得部によって取得された人口情報と、前記モデル演算部によって得られた出力とを比較して、前記エリアの人口の状態を判定する判定部と、
前記判定部による判定に用いられる判定基準を生成する判定基準生成部と、を備え、
前記判定基準生成部は、判定基準生成用の人口情報であって、同一エリアの第1の状態及び第1の状態とは異なる第2の状態それぞれでの判定基準生成用の第1の人口情報、並びに判定対象となるエリアの第1の状態での判定基準生成用の第2の人口情報を取得し、取得した第1及び第2の人口情報を、予め記憶した判定基準生成用のエンコーダ・デコーダモデルに入力して演算を行って、当該エンコーダ・デコーダモデルからの出力を得て、当該エンコーダ・デコーダモデルへの入力と出力とを比較して、比較結果に基づいて判定基準を生成する人口状態判定システム。
[2]前記第1の状態は平常時であり、前記第2の状態は異常時である[1]に記載の人口状態判定システム。
[3]前記判定部は、前記判定基準として閾値を用いて、人口が平常時とは異なる異常な状態であるか否かを判定し、
前記判定基準生成部は、前記閾値を生成する[2]に記載の人口状態判定システム。
[4]前記判定基準生成部は、第1の人口情報についての第1の状態及び第2の状態のそれぞれでの比較結果に基づく値の比、第2の人口情報についての比較結果に基づく値から前記閾値を生成する[3]に記載の人口状態判定システム。
[5]前記判定基準生成部は、複数のエリアそれぞれの第1の人口情報を取得し、複数のエリアそれぞれの前記比較結果に基づく値の比の統計値から、前記閾値を生成する[4]に記載の人口状態判定システム。
[6]前記判定基準生成部は、第1の状態での判定基準生成用の第1の人口情報から機械学習によって生成された判定基準生成用のエンコーダ・デコーダモデルに、第1の人口情報を入力し、判定基準生成用の第2の人口情報から機械学習によって生成された判定基準生成用のエンコーダ・デコーダモデルに、第2の人口情報を入力する[1]~[5]の何れかに記載の人口状態判定システム。
[7]前記取得部は、人口の状態の判定対象となるエリアの種別を示す種別情報を取得し、
前記モデル演算部は、前記取得部によって取得された種別情報に基づいて、前記エンコーダ・デコーダモデルを用いた演算を行う、[1]~[6]の何れかに記載の人口状態判定システム。
[8]前記モデル演算部は、前記種別情報も前記エンコーダ・デコーダモデルに入力して、当該エンコーダ・デコーダモデルからの出力を得る[7]に記載の人口状態判定システム。
[9]前記モデル演算部は、前記種別情報に基づいて、予め記憶した複数のエンコーダ・デコーダモデルから演算に用いるエンコーダ・デコーダモデルを選択して、選択したエンコーダ・デコーダモデルを用いた演算を行う[7]又は[8]に記載の人口状態判定システム。
Claims (9)
- 人口の状態の判定対象となるエリアの時系列の人口を示す人口情報を取得する取得部と、
前記取得部によって取得された人口情報を、入力データを圧縮して復元する予め記憶したエンコーダ・デコーダモデルに入力して演算を行って、当該エンコーダ・デコーダモデルからの出力を得るモデル演算部と、
前記取得部によって取得された人口情報と、前記モデル演算部によって得られた出力とを比較して、前記エリアの人口の状態を判定する判定部と、
前記判定部による判定に用いられる判定基準を生成する判定基準生成部と、を備え、
前記判定基準生成部は、判定基準生成用の人口情報であって、同一エリアの第1の状態及び第1の状態とは異なる第2の状態それぞれでの判定基準生成用の第1の人口情報、並びに判定対象となるエリアの第1の状態での判定基準生成用の第2の人口情報を取得し、取得した第1及び第2の人口情報を、予め記憶した判定基準生成用のエンコーダ・デコーダモデルに入力して演算を行って、当該エンコーダ・デコーダモデルからの出力を得て、当該エンコーダ・デコーダモデルへの入力と出力とを比較して、比較結果に基づいて判定基準を生成する人口状態判定システム。 - 前記第1の状態は平常時であり、前記第2の状態は異常時である請求項1に記載の人口状態判定システム。
- 前記判定部は、前記判定基準として閾値を用いて、人口が平常時とは異なる異常な状態であるか否かを判定し、
前記判定基準生成部は、前記閾値を生成する請求項2に記載の人口状態判定システム。 - 前記判定基準生成部は、第1の人口情報についての第1の状態及び第2の状態のそれぞれでの比較結果に基づく値の比、第2の人口情報についての比較結果に基づく値から前記閾値を生成する請求項3に記載の人口状態判定システム。
- 前記判定基準生成部は、複数のエリアそれぞれの第1の人口情報を取得し、複数のエリアそれぞれの前記比較結果に基づく値の比の統計値から、前記閾値を生成する請求項4に記載の人口状態判定システム。
- 前記判定基準生成部は、第1の状態での判定基準生成用の第1の人口情報から機械学習によって生成された判定基準生成用のエンコーダ・デコーダモデルに、第1の人口情報を入力し、判定基準生成用の第2の人口情報から機械学習によって生成された判定基準生成用のエンコーダ・デコーダモデルに、第2の人口情報を入力する請求項1に記載の人口状態判定システム。
- 前記取得部は、人口の状態の判定対象となるエリアの種別を示す種別情報を取得し、
前記モデル演算部は、前記取得部によって取得された種別情報に基づいて、前記エンコーダ・デコーダモデルを用いた演算を行う、請求項1に記載の人口状態判定システム。 - 前記モデル演算部は、前記種別情報も前記エンコーダ・デコーダモデルに入力して、当該エンコーダ・デコーダモデルからの出力を得る請求項7に記載の人口状態判定システム。
- 前記モデル演算部は、前記種別情報に基づいて、予め記憶した複数のエンコーダ・デコーダモデルから演算に用いるエンコーダ・デコーダモデルを選択して、選択したエンコーダ・デコーダモデルを用いた演算を行う請求項7に記載の人口状態判定システム。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US18/872,887 US20250356168A1 (en) | 2022-07-27 | 2023-04-28 | Population state determination system |
| JP2024536788A JP7734852B2 (ja) | 2022-07-27 | 2023-04-28 | 人口状態判定システム |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022119460 | 2022-07-27 | ||
| JP2022-119460 | 2022-07-27 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024024201A1 true WO2024024201A1 (ja) | 2024-02-01 |
Family
ID=89705957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2023/016969 Ceased WO2024024201A1 (ja) | 2022-07-27 | 2023-04-28 | 人口状態判定システム |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US20250356168A1 (ja) |
| JP (1) | JP7734852B2 (ja) |
| WO (1) | WO2024024201A1 (ja) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020123011A (ja) * | 2019-01-29 | 2020-08-13 | Kddi株式会社 | 所定圏における滞在圏人口を推定するプログラム、装置及び方法 |
| JP2021177284A (ja) * | 2020-05-07 | 2021-11-11 | Kddi株式会社 | 複数の投稿時系列データを用いた異常・変化推定方法、プログラム及び装置 |
-
2023
- 2023-04-28 WO PCT/JP2023/016969 patent/WO2024024201A1/ja not_active Ceased
- 2023-04-28 JP JP2024536788A patent/JP7734852B2/ja active Active
- 2023-04-28 US US18/872,887 patent/US20250356168A1/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020123011A (ja) * | 2019-01-29 | 2020-08-13 | Kddi株式会社 | 所定圏における滞在圏人口を推定するプログラム、装置及び方法 |
| JP2021177284A (ja) * | 2020-05-07 | 2021-11-11 | Kddi株式会社 | 複数の投稿時系列データを用いた異常・変化推定方法、プログラム及び装置 |
Non-Patent Citations (1)
| Title |
|---|
| OCHIAI, KEIICHI: "Consideration of detecting non-designated evacuation centers by extracting populated areas during disasters", IPSJ SIG TECHNICAL REPORT, INFORMATION PROCESSING SOCIETY OF JAPAN, vol. 2021-HCI-195, no. 18, 23 November 2021 (2021-11-23), pages 1 - 6, XP009552789, ISSN: 2188-8698 * |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2024024201A1 (ja) | 2024-02-01 |
| US20250356168A1 (en) | 2025-11-20 |
| JP7734852B2 (ja) | 2025-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113852603B (zh) | 网络流量的异常检测方法、装置、电子设备和可读介质 | |
| CN111612038B (zh) | 异常用户检测方法及装置、存储介质、电子设备 | |
| WO2018207878A1 (ja) | 需要予測装置 | |
| JP7524168B2 (ja) | 特徴抽出装置及び状態推定システム | |
| US11202175B2 (en) | At-home prediction device | |
| JP7692496B2 (ja) | 人口状態判定システム及びモデル生成システム | |
| WO2024024201A1 (ja) | 人口状態判定システム | |
| US20230214716A1 (en) | Learning device | |
| JP7377678B2 (ja) | 異常検知モデル学習装置、異常検知モデル及び異常検知装置 | |
| JP7522102B2 (ja) | 状況推定装置、状況推定プログラム、推定モデル及び状況推定方法 | |
| JPWO2019167685A1 (ja) | 訪問時刻決定装置 | |
| US20230169414A1 (en) | Population extraction device | |
| US20240386499A1 (en) | Investment trust price prediction device and prediction model | |
| JP7705861B2 (ja) | レコメンドシステム | |
| US11748424B2 (en) | Visiting destination prediction device and visiting destination prediction method | |
| WO2025052743A1 (ja) | 人口状態判定システム及びモデル生成システム | |
| JP2018181008A (ja) | サーバ装置 | |
| US20210123765A1 (en) | Pastime preference estimation device and pastime preference estimation method | |
| US20250095030A1 (en) | Recommendation system | |
| JP7513988B2 (ja) | テキスト入力タスクに基づいたタスク実行抑制機能の推定システム | |
| WO2024048036A1 (ja) | 店舗判定装置 | |
| US20230289863A1 (en) | Recommendation system | |
| JP2019028687A (ja) | サーバ装置及びパンデミック予測方法 | |
| JP7809822B2 (ja) | 店舗判定装置 | |
| JP2021099701A (ja) | リスク推定装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23845940 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2024536788 Country of ref document: JP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 18872887 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 23845940 Country of ref document: EP Kind code of ref document: A1 |
|
| WWP | Wipo information: published in national office |
Ref document number: 18872887 Country of ref document: US |

