JP2018205956A - Regional characteristic prediction method, regional characteristic prediction device, and regional characteristic prediction program - Google Patents

Abstract

To enable the prediction of regional characteristics with high accuracy.SOLUTION: A spatial weight matrix generation part generates a spatial weight matrix showing an adjacent relation between subdivided regions (meshes) with reference to a mesh map DB (S16), and a correction part corrects the spatial weight matrix with reference to a route DB (S18 to S30). Then, a prediction part predicts characteristics of a region on the basis of the corrected spatial weight matrix and attribute data. Thus, the characteristics of the region are predicted from the attribute data by taking both the adjacent relation between the meshes and influence (asymmetry) on connection between the adjacent meshes by a route into consideration.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an area characteristic prediction method, an area characteristic prediction apparatus, and an area characteristic prediction program.

  When predicting temporal changes in a region, such as changes in the flow of people or nature in the region, the influence of surrounding regions cannot be ignored as the target region becomes smaller. Therefore, conventionally, a method using a spatial weight matrix is known in metric statistics.

JP 2012-141953 A

  Some of the spatial weight matrices are determined by whether or not an area is adjacent, and some are determined by whether or not the area is within a certain distance range. The influence is expressed isotropically in the space weight matrix.

  However, even if it is adjacent to a certain area, each adjacent area is not necessarily affected to the same extent. Therefore, even if a spatial weight matrix in which the influence of adjacent areas is expressed isotropically, there is a possibility that temporal changes in the areas cannot be accurately predicted.

  An object of the present invention is to provide an area characteristic prediction method, an area characteristic prediction apparatus, and an area characteristic prediction program capable of accurately predicting an area characteristic.

  In one aspect, the regional characteristic prediction method refers to the first storage unit that stores the map information, generates a spatial weight matrix indicating the adjacent relationship between the subdivided regions included in the map information, and The spatial weight matrix is corrected by referring to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with information, the corrected spatial weight matrix, and a map The computer executes processing for predicting the characteristics of the prediction target based on the data related to the characteristics of the prediction target associated with the information.

  As one aspect, regional characteristics can be accurately predicted.

It is a figure which shows roughly the hardware constitutions of the regional characteristic prediction apparatus which concerns on 1st Embodiment. It is a functional block diagram of an area characteristic prediction device. FIG. 3A is a diagram illustrating an example of mesh data included in the mesh map DB, and FIG. 3B is a diagram illustrating an example of center coordinates of the mesh. FIG. 4A is a diagram showing route overlap mesh data included in the route DB, and FIG. 4B is a diagram showing route data included in the route DB. It is a figure which shows an example of attribute data. It is a flowchart which shows the process of an area characteristic prediction apparatus. FIG. 7A is a diagram illustrating nine meshes used in the description of FIG. 6, and FIG. 7B is a diagram illustrating a spatial weight matrix in the mesh of FIG. 7A. FIG. 8A is a diagram showing the route overlay mesh data used for the description of FIG. 6, and FIG. 8B is a diagram showing the spatial weight matrix corrected using FIG. 8A. FIG. 9A is a graph showing the prediction result in the comparative example, and FIG. 9B is a graph showing the prediction result in the first embodiment. It is a graph which shows a measured value. It is a table | surface which shows the result of having compared FIG. 10 with FIG. 9 (a) and FIG.9 (b). It is a figure which shows the mesh data with which the river used by 2nd Embodiment was piled up. FIG. 13A is a graph showing the prediction result in the comparative example, and FIG. 13B is a graph showing the prediction result in the second embodiment. It is a graph which shows the actual measurement value in a 2nd embodiment. It is a figure which shows a modification.

<< First Embodiment >>
Hereinafter, the first embodiment of the regional characteristic prediction apparatus will be described in detail with reference to FIGS.

  FIG. 1 schematically shows a hardware configuration of the regional characteristic prediction apparatus 10. The regional characteristic prediction apparatus 10 is, for example, a PC (Personal Computer), and as shown in FIG. 1, a CPU (Central Processing Unit) 90, a ROM (Read Only Memory) 92, a RAM (Random Access Memory) 94, a storage unit ( Here, an HDD (Hard Disk Drive) 96, a network interface 97, a display unit 93, an input unit 95, a portable storage medium drive 99, and the like are provided. Each component of the regional characteristic prediction apparatus 10 is connected to a bus 98. In the regional characteristic prediction apparatus 10, a program (including a regional characteristic prediction program) stored in the ROM 92 or the HDD 96, or a program (including a regional characteristic prediction program) read from the portable storage medium 91 by the portable storage medium drive 99. ) Is executed by the CPU 90, the functions of the respective units shown in FIG. 2 are realized. In FIG. 2, various DBs and data stored in the HDD 96 and the like are also illustrated.

  FIG. 2 is a functional block diagram illustrating functions realized by the CPU 90 of the regional characteristic prediction apparatus 10 executing a program. As illustrated in FIG. 2, the functions of the space weight matrix generation unit 12, the correction unit 14, and the prediction unit 16 as the generation unit are realized by the CPU 90 executing the program.

  The space weight matrix generation unit 12 generates a space weight matrix with reference to mesh data stored in the mesh map DB 22 as the first storage unit. Here, mesh data as shown in FIG. 3A is stored in the mesh map DB 22. The mesh data is information on a rectangular mesh arranged on the map, for example, 1 km in length and width, and includes the size of each mesh and the center coordinates (X, Y) of each mesh. Note that the areas included in the map are subdivided into a plurality of areas by each mesh. The space weight matrix generation unit 12 acquires the center coordinates of each mesh as shown in FIG. 3B from the mesh data. Then, based on the distance between the central coordinates, the adjacent relationship of each mesh is specified, and a spatial weight matrix is generated based on the specified adjacent relationship (whether or not adjacent).

  The correction unit 14 refers to the route DB 24 as the second storage unit and corrects the space weight matrix. Here, the route DB 24 includes route overlay mesh data as shown in FIG. 4A and route data as shown in FIG. 4B. The route overlay mesh data is data obtained by superimposing the mesh data and the layout of the route, as shown in FIG. In FIG. 4A, two routes are superimposed on the mesh data. The route data is data generated from the route overlay mesh data in FIG. 4A, and has a “route name” field and a “mesh ID” field as shown in FIG. 4B. Yes. In the route data, the name of the route and the identifier (mesh ID) of the mesh through which the route passes are associated.

  When the same route passes through two adjacent meshes, the correction unit 14 corrects the elements of the space weight matrix corresponding to these two adjacent meshes. For example, the elements of the spatial weight matrix corresponding to two meshes are multiplied by α (α is a value larger than 1). By performing such correction, the size (strength) between the meshes due to the presence of the route can be expressed in the space weight matrix.

  Based on the spatial weight matrix corrected by the correction unit 14 and the attribute data 26, the prediction unit 16 predicts data regarding characteristics of each region (for example, the number of single households in each mesh). Here, in FIG. 5, as an example of the attribute data 26, data on the number of single households of each mesh in each year is shown. The prediction unit 16 predicts the number of single households in each mesh of the next year, for example, using the attribute data 26 of FIG. Note that the attribute data 26 can be said to be data in which information related to characteristics predicted by the prediction unit 16 (characteristics to be predicted) is associated with map information (each mesh).

(Regarding the processing of the regional characteristic prediction device 10)
Next, the processing of the regional characteristic prediction apparatus 10 will be described in detail with reference to other drawings as appropriate along the flowchart of FIG. As an assumption that the processing of FIG. 6 is performed, it is assumed that the mesh map DB 22 and the route DB 24 store information related to regions where characteristics are to be predicted. Also, it is assumed that the attribute data 26 has already been input by the user via the input unit 95 or is included in the mesh map DB 22 as a dbf file. In the description of this processing, for simplification of description, the mesh data is data as shown in FIG. 7A, and the route overlay mesh data is data as shown in FIG. 8A. To do.

  In the process of FIG. 6, first, in step S12, the space weight matrix generation unit 12 acquires the center coordinates of the mesh. Specifically, the mesh ID that is the identifier of each mesh and the center coordinates (X, Y) are acquired from the mesh data of FIG. 7A (see FIG. 3B).

  Next, in step S14, the space weight matrix generation unit 12 determines an adjacency relationship between meshes. In this embodiment, it is determined that two meshes are adjacent when the distance between the center coordinates of the two meshes matches the size of the mesh. For example, when there are nine meshes 1 to 9 as shown in FIG. 7A, the meshes adjacent to the mesh 5 are 2, 4, 6, and 8. Further, the meshes adjacent to the mesh 1 are 2 and 4. In the above description, the space weight matrix generation unit 12 has described the case where the adjacency relationship between the meshes is determined based on the distance between the center coordinates of the two meshes, but is not limited thereto. For example, the space weight matrix generation unit 12 may determine whether two meshes are in contact with each other, and may determine that they are adjacent when they are in contact with each other with a line.

  Next, in step S16, the space weight matrix generation unit 12 generates a space weight matrix. Specifically, the space weight matrix generation unit 12 represents the adjacent relationship between meshes determined in step S14 as a matrix. In the case of FIG. 7A, the space weight matrix generation unit 12 generates a space weight matrix as shown in FIG. In the spatial weight matrix of FIG. 7B, for example, for row “1”, the values of the matrix elements in the columns “2” and “4” of mesh numbers adjacent to mesh 1 are set to 1, and the others are set to 0. It is said. For example, for row “2”, the values of matrix elements in the columns “1”, “3”, and “5” of mesh numbers adjacent to mesh 2 are set to 1, and the others are set to 0. Hereinafter, the values of the respective matrix elements are set in the same manner.

  When the process of step S16 ends, the process of correcting the space weight matrix is executed by the correction unit 14 in subsequent steps S18 to S30.

  In step S18, the correction unit 14 specifies an unspecified mesh. For example, it is assumed that the correction unit 14 specifies the mesh 1 in FIG.

  Next, in step S20, the correction unit 14 determines whether or not the route passes through the identified mesh. In this case, the correction unit 14 refers to the route overlap mesh data in FIG. 8A and determines whether or not the route passes through the mesh 1. In the example of FIG. 8A, since the route does not pass through the identified mesh 1, the determination in step S20 is denied, and the process proceeds to step S30. If transfering it to step S30, the correction | amendment part 14 will judge whether all the meshes have been specified. If the determination in step S30 is negative, the process returns to step S18. When returning to step S18, an unspecified mesh is specified. Here, for example, the mesh 2 is specified.

  Next, in step S20, it is determined whether or not a route passes through the identified mesh 2. In the example of FIG. 8A, since the route passes through the mesh 2, the determination in step S20 is affirmed, and the process proceeds to step S22.

  In step S22, the correcting unit 14 selects an unselected adjacent mesh. Here, one of the meshes 1, 3, and 5 adjacent to the mesh 2 (for example, mesh 1) is selected.

  Next, in step S24, the correction unit 14 determines whether or not the same route passes through the selected adjacent mesh. Here, since the same route as the mesh 2 specified in step S18 does not pass through the adjacent mesh 1 currently selected, the determination in step S24 is denied and the process proceeds to step S28. In step S28, the correction unit 14 determines whether or not all adjacent meshes have been selected. However, since only the mesh 1 of the meshes 1, 3, and 5 has been selected here, the determination is denied. Return to step S22.

  Returning to step S22, the correction unit 14 selects, for example, the mesh 3 as an unselected adjacent mesh. Next, in step S24, the correction unit 14 determines whether the same route as the mesh 2 specified in step S18 passes through the selected adjacent mesh 3. Here, the same line as the mesh 2 passes through the adjacent mesh 3 selected at the present time. Therefore, in this case, the determination in step S24 is affirmed, and the process proceeds to step S26.

  When the process proceeds to step S26, the correction unit 14 adds α (α> 1) to the corresponding matrix element. That is, the correction unit 14 sets α (= 1 × α) as the matrix element “1” in the second row from the top and the third column from the left in FIG. 7B (see FIG. 8B).

  Next, in step S28, the correction unit 14 determines whether or not all adjacent meshes have been selected, but here, only the meshes 1 and 3 of the meshes 1, 3 and 5 have been selected. The determination is denied and the process returns to step S22.

  Returning to step S22, the correction unit 14 selects, for example, the mesh 5 as an unselected adjacent mesh. Next, in step S <b> 24, the correction unit 14 determines whether or not the same route as the mesh 2 passes through the selected adjacent mesh 5. Here, the same line as the mesh 2 passes through the adjacent mesh 5 selected at the present time. Therefore, in this case, the determination in step S24 is affirmed, and the process proceeds to step S26.

  When the process proceeds to step S26, the correction unit 14 adds α (α> 1) to the corresponding matrix element. That is, the correction unit 14 sets α (= 1 × α) as the matrix element “1” in the second row from the top and the fifth column from the left in FIG. 7B (see FIG. 8B).

  Thereafter, the process proceeds to step S28, but since all adjacent meshes have been selected, the determination in step S28 is affirmed and the process proceeds to step S30. When the process proceeds to step S30, as described above, the correction unit 14 determines whether or not all the meshes have been specified. If the determination here is negative, the process returns to step S18.

  Thereafter, the correction unit 14 sequentially identifies the meshes 3 to 9 (S18), sequentially selects the meshes adjacent to the identified meshes (S22), and the same route passes through the identified meshes and the selected adjacent meshes. It is determined whether or not (S24). If the same route passes, α is added to the corresponding matrix element (S26) to correct the space weight matrix.

  In the case of the example in FIG. 8A, the space weight matrix is corrected as shown in FIG. 8B.

  As described above, when all the meshes have been specified and the determination in step S30 is affirmed, the process proceeds to step S32, and the prediction unit 16 uses the attribute data 26 and the corrected spatial weight matrix to determine the regional characteristics. Predict. For example, the prediction unit 16 has a function of performing Bayesian estimation using a programming language for statistical modeling (for example, Stan), and predicts regional characteristics based on the attribute data 26 and the corrected spatial weight matrix. Execute. For example, when the attribute data 26 is data indicating the transition of the number of single households in each mesh in the past, the prediction unit 16 determines, for example, the next year in each mesh from the attribute data 26 and the corrected space weight matrix. Predict the number of single households. In this case, when the same route passes between two adjacent meshes, since the value of the matrix element of the space weight matrix corresponding to the two meshes is largely corrected, the prediction unit 16 performs the correction after the correction. By using the spatial weight matrix, it is possible to predict regional characteristics in consideration of the influence between meshes due to the presence of the route (for example, the distribution of people tends to spread along the route). .

  As described above, when the processing up to step S32 is completed, the entire processing in FIG. 6 is completed.

  In addition, as shown to Fig.4 (a), the some route may pass in the some mesh of analysis object. In such a case, it is only necessary to repeat the processing of steps S18 to S30 in FIG. At this time, α may be different for each route. For example, α may be increased for a conventional line, and α may be decreased for a route that is unlikely to affect between regions such as the Shinkansen.

  Fig. 9 (a) shows data from 1995 (2005) to 2005 (2005) of the number of single households in each mesh set in a municipality as shown in Fig. 3 (a) (Fig. 5 The result of predicting the number of single households in each mesh in 2010 using the attribute data 26) and the space weight matrix (without correction) (predicted results only in 2010, others are attribute data 26) The actual measurement values included) are shown. Also, in FIG. 9B, the number of single households in each mesh in 2010 (2010) is shown in the same manner as in FIG. 9A by using the spatial weight matrix corrected as in the first embodiment. The predicted results are shown. Further, FIG. 10 shows actual measurement values of the number of single households for each mesh.

  FIG. 11 shows the difference (difference: comparison example) between the predicted value in FIG. 9A and the actually measured value in FIG. 10, and the difference (difference: difference) between the predicted value in FIG. 9B and the actually measured value in FIG. Application). From the table of FIG. 11, the difference: the average of the absolute values of the comparative examples was found to be 78.8, and the difference: the average of the absolute values of the present application was found to be 27.0. As a result, it was found that the number of single households can be accurately predicted by using the corrected spatial weight matrix as in the first embodiment.

  As described above in detail, according to the first embodiment, the space weight matrix generating unit 12 refers to the mesh map DB 22 and indicates the space weight matrix indicating the adjacent relationship between the subdivided regions (mesh). (S16), the correction unit 14 refers to the route DB 24 and corrects the space weight matrix (S18 to S30). Then, the prediction unit 16 predicts regional characteristics based on the corrected space weight matrix and attribute data. Accordingly, in the first embodiment, regional characteristics can be predicted from attribute data in consideration of not only the adjacency relationship between meshes but also the influence (asymmetry) of routes on the connection between adjacent meshes. it can. Therefore, it is possible to accurately predict regional characteristics.

  In the first embodiment, since the spatial weight matrix is corrected so that the connection between the meshes through which the route passes becomes strong, that is, the weight between the meshes through which the route passes is increased, the distribution of people is along the route. In consideration of the tendency to spread, it is possible to accurately predict regional characteristics.

  In the first embodiment, the case where the matrix elements are corrected based on whether or not the same route passes between the meshes has been described. A matrix element may be corrected based on whether a route, a main road, or the like) passes between meshes. In addition, the matrix elements may be corrected based on a plurality of traffic networks. In this case, the correction value (α) may be changed for each type of traffic network. In addition, the tendency of the distribution of people along the route is higher in the city center, and the tendency of the distribution of people along the main road is higher in the suburbs. The value (α) may be adjusted.

  For example, when the length of a route that passes through a mesh is specified and the accuracy of the prediction result of a mesh that passes only slightly is low, the length of the route is set to increase the accuracy of the prediction result. Α may be adjusted accordingly. Further, the correction may not be performed when the length of the route passing through the mesh is less than a predetermined length.

<< Second Embodiment >>
Next, a second embodiment will be described with reference to FIGS.

  In the first embodiment, the case where the spatial weight matrix is corrected so as to strengthen the connection of adjacent meshes that pass through the same route has been described. In the second embodiment, the region is divided by the river. An example in which the spatial weight matrix is corrected in consideration of the influence of (elements that hinder traffic) will be described.

  In the second embodiment, when a river flows between two adjacent meshes, the correction unit 14 multiplies the values of matrix elements corresponding to the two meshes by β (β <1). to correct. For example, in the mesh data in which rivers are superimposed as shown in FIG. 12, the meshes adjacent to mesh 5 are 2, 4, 6, and 8, but there are rivers between mesh 5 and meshes 2 and 6. The mesh is divided. Therefore, the values of the matrix elements corresponding to the meshes 5 and 2 and the meshes 5 and 6 are multiplied by β. The other meshes are similarly corrected. Then, the prediction unit 16 predicts regional characteristics (for example, the number of unemployed persons) using the corrected space weight matrix.

  Figure 13 (a) shows complete unemployment in 2010 using data on the number of unemployed persons (from 1995 to 2005) in each mesh set for a certain local government and a spatial weight matrix (not corrected). The result (comparative example) of predicting the number of persons is shown. FIG. 13B shows the result of predicting the number of completely unemployed people in 2010 using the same data as in FIG. 13A and the corrected spatial weight matrix (second implementation). Form). Further, FIG. 14 shows the transition (actual value) of the number of completely unemployed persons from 1995 to 2010.

  From these results, as in the first embodiment, when the average value of the absolute values of the difference between the predicted value of the comparative example and the actually measured value is obtained, it becomes 10.0, and the predicted value and the actually measured value of the second embodiment are calculated. The average value of the absolute values of the differences was 9.6. That is, the prediction accuracy was improved by 4% by correcting the space weight matrix as in the second embodiment.

  As described above, according to the second embodiment, when information on an element (for example, a river) that hinders traffic between meshes is used as an element that affects the connection between meshes, and there is a river between meshes Since the matrix elements are corrected so as to weaken the connection between the meshes, the characteristics of the region can be accurately predicted in consideration of the elements that obstruct the traffic between the meshes.

  In the above embodiment, the case where the number of completely unemployed persons is predicted as a regional characteristic has been described. However, the present invention is not limited to this. Good.

  In the above embodiment, the case where the element that hinders the traffic between the meshes is a river has been described as an example. There may be.

  In the first and second embodiments, the description has been given of the case where the prediction unit 16 predicts the regional characteristics by Bayesian estimation. However, the present invention is not limited to this, and the regional characteristics may be predicted by other methods. Good.

  Note that the first and second embodiments may be combined. In other words, when the traffic network connects between meshes, the matrix element value is corrected to be large, and when there are elements that obstruct traffic such as rivers between meshes, the matrix element value is small. You may correct to.

  In addition, although the said 1st, 2nd embodiment demonstrated the case where the area characteristic prediction apparatus 10 had each function of FIG. 2, it is not restricted to this. For example, the cloud server 110 connected to a network 80 such as the Internet as shown in FIG. In this case, the cloud server 110 may receive the data input from the user terminal 70, and the cloud server 110 may execute the process of FIG. Then, the processing result (prediction result) is transmitted from the cloud server 110 to the user terminal 70, and the processing result received at the user terminal 70 is used. The cloud server 110 may be installed either in Japan or overseas.

  The above processing functions can be realized by a computer. In that case, a program describing the processing contents of the functions that the processing apparatus should have is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium (except for a carrier wave).

  When the program is distributed, for example, it is sold in the form of a portable recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

  The above-described embodiment is an example of a preferred embodiment of the present invention. However, the present invention is not limited to this, and various modifications can be made without departing from the scope of the present invention.

In addition, regarding the above description of the first and second embodiments, the following additional notes are disclosed.
(Additional remark 1) With reference to the 1st storage part which stores map information, the space weight matrix which shows the adjacency relation between the segmented area contained in the said map information is generated,
Referring to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information, and corrects the spatial weight matrix;
Predicting the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction method characterized in that a computer executes processing.
(Supplementary note 2) The regional characteristic prediction method according to supplementary note 1, wherein information on an element that influences the connection between the subdivided regions includes information on a traffic network.
(Supplementary note 3) The regional characteristic prediction method according to supplementary note 2, wherein, in the correction process, the spatial weight matrix is corrected so that a weight between regions through which the traffic network passes is increased.
(Supplementary Note 4) The element information that affects the connection between the subdivided regions includes information on an element that inhibits traffic between the regions. Regional characteristic prediction method.
(Supplementary note 5) The regional characteristic prediction method according to supplementary note 4, wherein, in the correction process, the spatial weight matrix is corrected so that a weight between regions where traffic is inhibited is small.
(Additional remark 6) With reference to the 1st storage part which stores map information, the production | generation part which produces | generates the space weight matrix which shows the adjacent relationship between the subdivided areas contained in the said map information,
A correction unit that corrects the spatial weight matrix with reference to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information;
A prediction unit that predicts the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction apparatus comprising:
(Supplementary note 7) The regional characteristic prediction apparatus according to supplementary note 6, wherein the information on the elements that influence the connection between the subdivided regions includes information on a traffic network.
(Additional remark 8) The said correction | amendment part correct | amends the said spatial weight matrix so that the weight between the areas which the said traffic network passes may increase, The regional characteristic prediction apparatus of Additional remark 7 characterized by the above-mentioned.
(Additional remark 9) The information of the element which influences the coupling | bonding between the said subdivided areas contains the information of the element which inhibits the traffic between the said areas, The additional description 6-8 characterized by the above-mentioned Regional characteristic prediction device.
(Additional remark 10) The said correction | amendment part correct | amends the said spatial weight matrix so that the weight between the areas where traffic may be inhibited becomes small, The regional characteristic prediction apparatus of Additional remark 9 characterized by the above-mentioned.
(Additional remark 11) With reference to the 1st storage part which stores map information, the spatial weight matrix which shows the adjacent relationship between the subdivided area contained in the said map information is produced | generated,
Referring to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information, and corrects the spatial weight matrix;
Predicting the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction program for causing a computer to execute processing.

10 Regional Characteristic Prediction Device 12 Spatial Weight Matrix Generation Unit (Generation Unit)
14 correction | amendment part 16 prediction part 22 mesh map DB (1st storage part)
24 route DB (2nd storage part)
26 Attribute data (data related to the characteristics of the target)

Claims (7)

Referring to a first storage unit that stores map information, and generating a spatial weight matrix that indicates an adjacent relationship between subdivided regions included in the map information;
Referring to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information, and corrects the spatial weight matrix;
Predicting the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction method characterized in that a computer executes processing.   The regional characteristic prediction method according to claim 1, wherein the information on the elements that influence the connection between the subdivided regions includes information on a traffic network.   The regional characteristic prediction method according to claim 2, wherein in the correction process, the spatial weight matrix is corrected so that a weight between regions through which the traffic network passes is increased.   The region according to any one of claims 1 to 3, wherein the information on an element that affects the connection between the subdivided regions includes information on an element that inhibits traffic between the regions. Characteristic prediction method.   The regional characteristic prediction method according to claim 4, wherein in the correction process, the spatial weight matrix is corrected so that a weight between regions where traffic is hindered is reduced. A generating unit that generates a spatial weight matrix indicating an adjacent relationship between the subdivided regions included in the map information with reference to a first storage unit that stores map information;
A correction unit that corrects the spatial weight matrix with reference to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information;
A prediction unit that predicts the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction apparatus comprising: Referring to a first storage unit that stores map information, and generating a spatial weight matrix that indicates an adjacent relationship between subdivided regions included in the map information;
Referring to a second storage unit that stores information on elements that influence the connection between the subdivided regions in association with map information, and corrects the spatial weight matrix;
Predicting the characteristics of the prediction target based on the corrected spatial weight matrix and data on the characteristics of the prediction target associated with the map information;
A regional characteristic prediction program for causing a computer to execute processing.

Images