Evaluation method, evaluation system, and evaluation program

Abstract

To provide an evaluation method, an evaluation system, and an evaluation program capable of achieving appropriate evaluation for a business operator to select a candidate area suitable for a location, from among a plurality of candidate areas included in a public space.SOLUTION: A method of evaluating a candidate area, for a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space includes: extracting movement histories with respect to destinations which are set according to a business operator, from action histories of persons who move in a public space in a predetermined period; acquiring, using the movement histories, for each person, a degree of interest in each of the destinations, based on a cost for moving to the destination within the predetermined period and the number of visits to the destination; acquiring, for each person, the number of times that each of the candidate areas is used within the predetermined period; and acquiring evaluation values of the candidate areas on the basis of the degrees of interest and the numbers of times of use.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an evaluation system, an evaluation program, and an evaluation method for selecting a candidate area suitable for locating a business facility for the purpose of constructing or relocating the facility from among a plurality of candidate areas. [Background technology]

[0002] Conventionally, in order to support marketing of stores located in public spaces, an information processing device has been known that uses an imaging device attached to a power distribution facility to acquire pedestrian information by analyzing pedestrians passing through the public space, and outputs evaluation information of stores located at each point in the public space (see Patent Document 1).

[0003] Also known in the art is a user behavior analysis device that performs evaluations taking into account the behavioral tendencies of station users, and calculates an index showing the behavioral tendencies of users who use transportation payment means for each station based on the entry / exit history and purchase history contained in the usage history of transportation payment means, and then calculates an index showing the characteristics of the station based on the index showing the behavioral tendencies (see Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6825613 [Patent Document 2] Patent No. 5638115 Summary of the Invention [Problem to be solved by the invention]

[0005] In the conventional technologies described in the above Patent Documents 1 and 2, store evaluation information and station indexes are obtained based on passerby information and entrance / exit history. Passerby information and entrance / exit history include information on passersby who are commuting to school or work, etc. Therefore, there is a problem that information on people who use stores is buried in information on passersby, making it difficult to use for marketing.

[0006] Therefore, the present disclosure aims to provide an evaluation method, evaluation system, and evaluation program that enable businesses to make appropriate evaluations to select candidate areas suitable for location from among multiple candidate areas included in public spaces. [Means for solving the problem]

[0007] The evaluation method disclosed herein is a method for evaluating candidate areas in order for a business to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, in which a processor implementing the evaluation method extracts a movement history to destinations set according to the business from the behavioral history of each person moving through the public space over a specified period of time, and uses the movement history to obtain, for each person, a degree of interest in each destination based on the cost of traveling to the destination within the specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0008] The evaluation method disclosed herein is a method for evaluating candidate areas, which allows a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, and a processor implementing the evaluation method executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the analysis step, the processor obtains, for each person, a level of interest for each destination set in accordance with the business operator, and obtains, for each person, a vector whose elements are the level of interest for each destination as an attribute vector indicating attributes related to the interests of the person, and clusters the people based on the attribute vector.

[0009] The evaluation method disclosed herein is a method for evaluating candidate areas, which allows a business operator to select a candidate area suitable for location from among a plurality of candidate areas contained in a public space, and a processor implementing the evaluation method executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the evaluation step, the processor obtains a representative value of the number of times each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator intends to locate for each of the classes, and performs the evaluation by weighting and adding up the representative values ​​of the number of times each of the candidate areas based on the representative value of the degree of interest.

[0010] The evaluation system disclosed herein is a candidate area evaluation system for enabling a business to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, and includes a processor for performing evaluation, wherein the processor extracts a movement history to destinations set according to the business from the behavioral history of each person moving through the public space, and uses the movement history to obtain, for each person, a degree of interest for each destination based on the cost of traveling to the destination within a specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0011] The evaluation system disclosed herein is a candidate area evaluation system for enabling a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, and includes a processor for performing evaluation, and the processor executes an analysis step of clustering people based on behavioral information of people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the analysis step, obtains, for each person, a level of interest for each destination set in accordance with the business, and obtains, for each person, a vector whose elements are the level of interest for each destination as an attribute vector indicating attributes related to the interests of the person, and performs clustering based on the attribute vector.

[0012] The evaluation system disclosed herein is a candidate area evaluation system for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas contained in a public space, and includes a processor for performing evaluation, wherein the processor executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the evaluation step, the processor obtains a representative value of the number of times each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator intends to locate for each of the classes, and performs the evaluation by weighting and adding up the representative values ​​of the number of times each of the candidate areas based on the representative value of the degree of interest.

[0013] The evaluation program disclosed herein is a candidate area evaluation program for enabling a business operator to select a candidate area suitable for location from among a number of candidate areas included in a public space, in which a processor for performing the evaluation extracts movement history to destinations set in accordance with the business operator from the behavioral history of each person moving through the public space, and uses the movement history to obtain, for each person, a degree of interest in each destination based on the cost of traveling to the destination within a specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0014] The evaluation program disclosed herein is a candidate area evaluation program for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, in which a processor for performing the evaluation executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, in which, in the analysis step, a level of interest for each destination set in accordance with the business is obtained for each person, and for each person, a vector whose elements are the level of interest for each destination is obtained as an attribute vector indicating attributes related to the interests of the person, and clustering is performed based on the attribute vector.

[0015] The evaluation program disclosed herein is a candidate area evaluation program for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, in which a processor for performing the evaluation executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the evaluation step, the processor obtains a representative value of the number of times each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator intends to locate for each of the classes, and performs the evaluation by weighting and adding up the representative values ​​of the number of times each of the candidate areas based on the representative value of the degree of interest. Effect of the Invention

[0016] According to the present disclosure, it is possible to provide an evaluation method, evaluation system, and evaluation program that can perform an appropriate evaluation to select a candidate area suitable for location from among multiple candidate areas included in a public space. [Brief description of the drawings]

[0017] [Figure 1]FIG. 1A is an overall configuration diagram of an evaluation system according to an embodiment, and FIG. 1B is an example of an image captured by an imaging device installed in a destination facility. [Diagram 2] Evaluation system block diagram [Diagram 3] Functional block diagram of the evaluation system [Figure 4] (A) Example of person identification table and (B) example of behavior history table [Diagram 5] Example of behavior history information transmitted to evaluation device [Figure 6] FIG. 13 is an explanatory diagram for explaining extraction of a destination facility and search for a departure point in action history information; [Figure 7] FIG. 13 is an explanatory diagram for explaining calculation of a cost to reach a destination facility from action history information; [Figure 8] (A) An example of an attribute table, (B) an example of a table showing average cost and number of visits, and (C) a modified example of the attribute table. [Figure 9] Schematic diagram showing clustering by person vectors [Figure 10] Example of a usage count table [Figure 11] Examples of (A) interest level table, (B) usage count table, (C) evaluation value table, and (D) normal evaluation value table. [Figure 12] Example of candidate area input screen displayed on device [Figure 13] Example of the standby screen displayed on the device [Figure 14] Example of the results displayed on the device [Figure 15] A sequence diagram showing the flow of the evaluation process until the evaluation value (regular evaluation value) and candidate areas suitable for relocation are displayed. [Figure 16] FIG. 1 shows (A) an attribute table and (B) a table showing the average number of times a station is used according to an embodiment. [Figure 17] A distribution diagram showing the distribution of person vectors when the horizontal axis is the total cost for tenant a and the vertical axis is the total cost for tenant c. [Figure 18]1A is a diagram showing an interest level table, a usage frequency table, a rating value table, and a normal rating value table according to an embodiment of the present invention; [Figure 19] A modified example of the attribute table corresponding to FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] The first invention made to solve the above problem is a method for evaluating candidate areas, which allows a business operator to select a candidate area suitable for location from among a plurality of candidate areas contained in a public space, and is configured such that a processor implementing the evaluation method extracts a movement history to destinations set according to the business operator from the behavioral history of each person moving through the public space over a specified period of time, and uses the movement history to obtain, for each person, a degree of interest for each destination based on the cost of moving to the destination within the specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0019] According to this, by using the cost that a person spends until arriving at a destination and the number of visits to the destination, it is possible to estimate the degree of interest of each destination according to the person's business. As a result, by using the degree of interest and the number of times the candidate area is used, an evaluation value can be obtained that corresponds to the expected degree of use when a business is located in the candidate area, and therefore it is possible to provide an evaluation method that can perform an appropriate evaluation to select a candidate area suitable for a location.

[0020] In addition, the second invention is configured such that the cost is set based on at least one of a travel distance, a travel time, and an expense required to reach the destination.

[0021] This allows the cost of traveling to a destination to be properly obtained.

[0022] In addition, a third invention is configured such that the processor extracts the destination from the behavioral history, estimates the departure point by obtaining the position of the person before the time of arrival at the destination, and obtains the cost.

[0023] This allows you to obtain the cost required to reach your destination.

[0024] In addition, a fourth invention is configured such that the processor performs clustering to divide the person into classes using the sum of the costs for each of the destinations, and obtains the evaluation value by adding up the product of a representative value of the sum of the costs for each of the classes and a representative value of the number of times of use for each of the classes.

[0025] According to this, the evaluation value is calculated by adding up the product of the representative value of the total cost and the representative value of the number of uses. Therefore, the number of uses of the candidate area is weighted and added up according to the degree of interest in the business of each class of business operator, and the evaluation value is calculated, so that the evaluation value can be obtained in a way that better reflects the degree of use expected when locating in the candidate area.

[0026] In addition, a fifth invention is configured such that the processor obtains the evaluation value by adding up the product of an average value of the sum of the costs for each of the classes and an average value of the number of times of use for each of the classes.

[0027] This makes it possible to easily obtain a representative value of the total cost and a representative value of the number of uses.

[0028] A sixth invention made to solve the above problem is a method for evaluating candidate areas for a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, wherein a processor implementing the evaluation method executes an analysis step of clustering people based on behavioral information of people moving through the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the analysis step, the processor obtains, for each person, a level of interest for each destination set in accordance with the business operator, and obtains, for each person, a vector whose elements are the level of interest for each destination as an attribute vector indicating attributes related to the interests of the person, and clusters the people based on the attribute vector.

[0029] This allows people to be classified according to their level of interest, which in turn allows candidate areas to be evaluated according to the level of interest in a facility in which a business operator is considering locating, making it possible to perform an appropriate evaluation to select candidate areas suitable for locating the facility.

[0030] The seventh invention made to solve the above problem is a method for evaluating candidate areas for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas contained in a public space, wherein a processor implementing the evaluation method executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the evaluation step, the processor obtains a representative value of the number of times each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator wishes to locate for each of the classes, and performs the evaluation by weighting and adding up the representative values ​​of the number of times each of the use based on the representative value of the degree of interest.

[0031] According to this, the level of interest in the business facilities of operators of each class is obtained by adding up the costs, and the number of times a candidate area is used is weighted and added up according to that level of interest to calculate an evaluation value, so that an evaluation value can be obtained that better reflects the degree of usage expected when locating in a candidate area.

[0032] The eighth invention made to solve the above problem is a candidate area evaluation system for a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, the system including a processor for performing evaluation, wherein the processor extracts a movement history to destinations set according to the business operator from the behavior history of each person moving through the public space, and uses the movement history to obtain, for each person, a degree of interest for each destination based on the cost of moving to the destination within a specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0033] According to this, by using the cost that a person spends until arriving at a destination and the number of visits to the destination, it is possible to estimate the degree of interest of each destination according to the person's business. As a result, by using the degree of interest and the number of times the candidate area is used, an evaluation value can be obtained that corresponds to the expected degree of use when a business is located in the candidate area, so it is possible to provide an evaluation system that can perform appropriate evaluations to select candidate areas suitable for location.

[0034] In addition, a ninth invention is configured such that the processor performs clustering to divide the person into classes using the sum of the costs for each of the destinations, and obtains the evaluation value by adding up the product of a representative value of the sum of the costs for each of the classes and a representative value of the number of times of use for each of the classes.

[0035] According to this, the evaluation value is calculated by adding up the product of the representative value of the total cost and the representative value of the number of uses. Therefore, the number of uses of the candidate area is weighted and added up according to the degree of interest in the business of each class of business operator, and the evaluation value is calculated, so that the evaluation value can be obtained in a way that better reflects the degree of use expected when locating in the candidate area.

[0036] A tenth invention made to solve the above problem is a candidate area evaluation system for a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, the system including a processor for performing evaluation, the processor executing an analysis step of clustering people based on behavioral information of people moving through the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, in which, in the analysis step, a degree of interest for each destination set in accordance with the business operator is obtained for each person, and a vector having elements of the degree of interest for each destination for each person is obtained as an attribute vector indicating attributes related to the interests of the person, and clustering is performed based on the attribute vector.

[0037] This allows people to be classified according to their level of interest, which in turn allows candidate areas to be evaluated according to the level of interest in a facility in which a business operator is considering locating, making it possible to perform an appropriate evaluation to select candidate areas suitable for locating the facility.

[0038] An eleventh invention made to solve the above problem is a candidate area evaluation system for a business operator to select a candidate area suitable for location from among a plurality of candidate areas contained in a public space, the system including a processor for performing the evaluation, the processor executing an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, the processor in the evaluation step obtaining a representative value of the number of times of use for each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator wishes to locate for each of the classes, and obtaining an evaluation value by weighting and adding up the representative values ​​of the number of times of use based on the representative value of the degree of interest.

[0039] According to this, the level of interest in the business facilities of operators of each class is obtained by adding up the costs, and the number of times a candidate area is used is weighted and added up according to that level of interest to calculate an evaluation value, so that an evaluation value can be obtained that better reflects the degree of usage expected when locating in a candidate area.

[0040] A twelfth invention made to solve the above problem is a candidate area evaluation program for a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, wherein a processor for performing the evaluation extracts a movement history to destinations set according to the business operator from the behavior history of each person moving through the public space, and uses the movement history to obtain, for each person, a degree of interest for each destination based on the cost of moving to the destination within a specified period and the number of visits to the destination, obtains, for each person, the number of times each candidate area has been used within the specified period, and obtains an evaluation value for each candidate area based on the interest degree and the number of times the destination has been used.

[0041] According to this, by using the cost that a person spends until arriving at a destination and the number of visits to the destination, the degree of interest of each destination corresponding to the business of the person can be estimated. As a result, by using the degree of interest and the number of times the candidate area is used, an evaluation value corresponding to the expected degree of use when a business is located in the candidate area can be obtained, so that an evaluation program can be provided that can perform appropriate evaluation to select a candidate area suitable for a location.

[0042] In addition, the 13th invention is configured such that the processor performs clustering to divide the person into classes using the sum of the costs for each of the destinations, and obtains the evaluation value by adding up the product of a representative value of the sum of the costs for each of the classes and a representative value of the number of times of use for each of the classes.

[0043] According to this, the evaluation value is calculated by adding up the product of the representative value of the total cost and the representative value of the number of uses. Therefore, the number of uses of the candidate area is weighted and added up according to the degree of interest in the business of each class of business operator, and the evaluation value is calculated, so that the evaluation value can be obtained in a way that better reflects the degree of use expected when locating in the candidate area.

[0044] A fourteenth invention made to solve the above problem is a candidate area evaluation program for a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, wherein a processor for performing the evaluation executes an analysis step of clustering people based on behavioral information of people moving through the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the analysis step, a degree of interest for each destination set in accordance with the business operator is obtained for each person, and a vector whose elements are the degree of interest for each destination is obtained for each person as an attribute vector indicating attributes related to the interests of the person, and clustering is performed based on the attribute vector.

[0045] This allows people to be classified according to their level of interest, which in turn allows candidate areas to be evaluated according to the level of interest in a facility in which a business operator is considering locating, making it possible to perform an appropriate evaluation to select candidate areas suitable for locating the facility.

[0046] A fifteenth invention made to solve the above problem is a candidate area evaluation program for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, wherein a processor for performing the evaluation executes an analysis step of clustering people based on behavioral information of the people moving in the public space, and an evaluation step of evaluating the candidate areas based on the results of the analysis step, and in the evaluation step, the processor obtains a representative value of the number of times each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the business operator intends to locate for each of the classes, and performs the evaluation by weighting and adding up the representative values ​​of the number of times each of the candidate areas based on the representative value of the degree of interest.

[0047] According to this, the level of interest in the business facilities of operators of each class is obtained by adding up the costs, and the number of times a candidate area is used is weighted and added up according to that level of interest to calculate an evaluation value, so that an evaluation value can be obtained that better reflects the degree of usage expected when locating in a candidate area.

[0048] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0049] The evaluation system 1 according to the present invention provides business operators with evaluation information that is useful when operating, installing, or relocating facilities by analyzing the behavior of people in public spaces such as stations, stores, and roads. Facilities operated by business operators or facilities that they intend to install or relocate (hereinafter, business facilities) may include, for example, large amusement facilities such as theme parks and amusement parks, large commercial facilities such as department stores and fashion buildings, various stores, art museums, entertainment facilities, activity facilities, movie theaters, accommodation facilities, bathing facilities, and other various facilities.

[0050] The evaluation system 1 is used to perform evaluations showing whether or not a business's location plan or advertisement distribution plan is effective by analyzing the behavior of people in a public space.

[0051] Here, we will explain an example in which the evaluation system 1 obtains from a business operator multiple areas (hereinafter referred to as candidate areas) that are candidates for relocation, obtains an evaluation value that indicates the degree to which the candidate area is suitable as an area for establishing a business facility based on the behavioral history of people in public spaces, and presents this to the business operator.

[0052] In this embodiment, the evaluation system 1 acquires candidate areas from the operator by selecting a station (or a bus stop). That is, each candidate area is associated with one station, and each candidate area is associated with the surrounding area of ​​the corresponding station.

[0053] In order to perform such evaluations, as shown in FIG. 1(A), the evaluation system 1 is equipped with a behavior history acquisition device 3 (behavior history acquisition unit) that collects behavior history (also called behavior log) of people in public spaces such as stations, stores, and roads, an evaluation device 5 (evaluation unit) that acquires and outputs evaluation information of a desired area based on the behavior history acquired by the behavior history acquisition device 3, and a terminal 7 that acquires multiple candidate areas from users of the evaluation system 1 and notifies the users of the evaluation results.

[0054] The evaluation system 1 may further include a management terminal (not shown) for managing the behavior history acquisition device 3 and the evaluation device 5. One of the terminals 7 may function as the management terminal, and the management terminal may function as the terminal 7 that notifies the user of the result.

[0055] The behavior history acquisition device 3 includes a plurality of imaging devices 11 (imaging units) and a behavior history server 13 (behavior history acquisition unit). In addition to the imaging devices 11 and the behavior history server 13, the behavior history acquisition device 3 may also include a gate management device 19 that authenticates the passage through an entrance / exit gate 15 arranged in a public space. The imaging devices 11, the behavior history server 13, and the gate management device 19 are each connected to be able to communicate with each other via a network 21 such as the Internet.

[0056] The imaging device 11 is installed in a public space and is composed of a camera (also called a monitor) that captures images of people in the public space. The imaging device 11 may be composed of a camera 11A (for example, a camera for acquiring congestion status, a security camera, etc.) installed at a station or on the roadside. Alternatively, the imaging device 11 may be composed of a camera 11B (a camera for monitoring visitors) installed at the entrance / exit or inside a tenant or store. The imaging device 11 may also be a face recognition camera (not shown) installed in a store, a vending machine, etc., for performing face recognition for payment.

[0057] When an automatic ticket gate is configured to perform face authentication, the imaging device 11 may be configured with a face authentication camera 11C provided at the ticket gate or around the ticket gate. When various stores, facilities, conference rooms, and buildings are configured to lock and unlock doors by face authentication, the imaging device 11 may be configured with a face authentication camera. The imaging device 11 may also be a face authentication camera (not shown) of an authentication device 23 that opens and closes gates at concert venues and event venues by face authentication. The imaging device 11 outputs the captured image to the action history server 13 via the network 21 together with the date and time of capture. The authentication method may be biometric authentication other than face authentication. Furthermore, other known authentication methods other than biometric authentication may be used as the authentication method.

[0058] The gate management device 19 reads information from a card held by a person, determines whether or not the person can pass through the entrance / exit gate 15, and opens and closes the gate. For example, the gate management device 19 may read information from a card held by a person (a so-called transportation IC card) and issue a command to a ticket gate to open and close the gate. The information read by the gate management device 19 includes an identification number for identifying the card. The gate management device 19 may transmit to the action history server 13 the time when the information (hereinafter, card information) was read from the card and the identification number corresponding to the card.

[0059] In order to classify a person, the evaluation system 1 needs to obtain the person's entry into a plurality of facilities including the business facilities of the person's business (hereinafter, destination facilities). In this embodiment, in order to confirm the person's entry into the destination facility, an image captured by an imaging device 11 installed at the entrance of the destination facility, card information read when the entrance / exit gate 15 is opened or closed, etc. are used.

[0060] A destination facility is a facility that is a destination to be visited by a person, and may be set as a single facility or multiple facilities depending on a business operator. The destination facility may be set by the operator of the evaluation system 1 in consideration of the business conducted by the business operator, or may be set based on input (selection) by the business operator.

[0061] The destination facilities preferably include facilities that are frequently visited by people who are interested in the business facility and facilities that are less frequently visited. It is preferable that the types of businesses and products handled by the destination facilities are different from each other.

[0062] The destination facility may be, for example, a store (tenant) that rents part of a large commercial facility such as a department store or fashion building and occupies the facility. In addition, the destination facility may be a small store, a street-front store such as a chain store such as a convenience store or drug store, an accommodation facility such as a hotel, or an entertainment facility such as a theme park or an amusement park. The destination facility may be a facility with a specific store name (for example, an ACE supermarket), or a facility that is not specific but is categorized to a certain extent (for example, a rich ramen shop). The following describes an example in which the destination facility is set to multiple tenants.

[0063] FIG. 1(B) shows a typical example of an image captured by an imaging device 11 installed in a destination facility. The captured image may be any image that can identify a person entering (entering) the destination facility. As shown in FIG. 1(B), when face recognition of a person is required, such as at a station ticket gate or when entering or exiting the destination facility, the captured image may include the face of a person entering (entering) the destination facility. In addition, when simply identifying a person, the captured image may include a part or the whole of the body of a person entering (entering) the destination facility.

[0064] 2, the behavior history server 13 may be configured by a computer equipped with known hardware such as a processor 31 (CPU, MPU, etc.), a memory 33 (RAM, ROM, etc.), a storage 35 (a storage device such as an HDD or SSD), and a communication interface 37. The behavior history server 13 may also be equipped with a display and input devices (a keyboard, a mouse, etc.).

[0065] As shown in FIG. 3, the behavior history server 13 includes a storage unit 41, a communication unit 43, a person identification unit 45, a behavior information acquisition unit 47, and a behavior information extraction unit 49.

[0066] The memory unit 41 is composed of the storage 35, and stores various information for identifying a person's behavior history. Specifically, the memory unit 41 stores an image captured by the imaging device 11 as image information together with the location and date and time when the image was captured. In addition, the memory unit 41 stores an identification number read from card information by the gate management device 19 as card reading information together with identification information for identifying the corresponding gate and the date and time when the image was read.

[0067] The information for identifying the behavior history stored in the storage unit 41 also includes person identification information for identifying a person included in an image acquired by the imaging device 11 within a predetermined period. The person identification information stores a person's identification information and the corresponding person's biometric information in association with each other. The biometric information may be any information that can identify a person from an image acquired by the imaging device 11, and may be, for example, a facial feature amount, a body feature amount, or the like.

[0068] FIG. 4(A) shows an example of person identification information according to the embodiment. The person identification information includes identification information, facial feature amount, and body feature amount of persons associated with each other. However, the present invention is not limited to this embodiment, and the person identification information may be any information for identifying a person, and may be composed of identification information and various biometric information of persons associated with each other. For example, the person identification information may be composed of identification information and facial feature amount of persons associated with each other, or may be composed of identification information and body feature amount of persons associated with each other. The person identification information is stored in the storage unit 41 as a table (hereinafter, person identification table T1).

[0069] The communication unit 43 includes a communication interface 37. The communication unit 43 performs wireless or wired communication with other external devices (such as the evaluation device 5 and the terminal 7) via the network 21 in accordance with a known communication protocol.

[0070] The person identification unit 45 uses the image information to identify a person reflected in the image. Specifically, the person identification unit 45 determines whether a person is reflected from the image stored in the storage unit 41 based on a known method. When a person is reflected, the person identification unit 45 extracts biometric information for identifying the person from the image. When the person identification information stored in the storage unit 41 contains information matching the extracted biometric information, the person identification unit 45 uses the person identification information to obtain identification information of the person reflected in the image. When the person identification information stored in the storage unit 41 does not contain information matching the extracted biometric information, the person identification unit 45 adds the biometric information to the person identification information together with the newly assigned person's identification information.

[0071] When the person identification unit 45 acquires identification information of a person appearing in an image, the behavioral information acquisition unit 47 uses the image information to acquire the location and date and time when the image was acquired, and associates the person's identification information with the location and date and time and stores them in the memory unit 41 as behavioral history information.

[0072] An example of the behavior history information is shown in Fig. 4(B). In the example shown in Fig. 4(B), the behavior history information is stored in the storage unit 41 in chronological order as a table (hereinafter, behavior history table T2) in which the identification information of a person is associated with the location and date and time when the person was photographed.

[0073] In addition, when authentication is performed by the gate management device 19, the behavior information acquisition unit 47 may extract facial features and body features of a person in an image captured by the imaging device 11 (camera 11D, see FIG. 1(A)) located around the entrance / exit gate 15 to acquire an identification number, and may associate the identification number with the identification number corresponding to the card that was read (also referred to as movement log integration). As a result, the behavior information unit may consider that the person holding the card was present at the location where the card was read, and record the corresponding person's identification number and the location and date and time when the card was read in the behavior history table T2 as behavior history information.

[0074] The behavioral information acquisition unit 47 holds behavioral history information for a certain period (hereinafter, accumulation period), and deletes (discards) the behavioral history information from the behavioral history table T2 after the certain period (hereinafter, accumulation period). The accumulation period is set, for example, on a monthly or annual basis, and may be configured so that the administrator of the behavioral history server 13 can set it.

[0075] The behavior information extraction unit 49 extracts corresponding behavior history information from the behavior history table T2 in which the behavior history information is recorded based on a request (query) from the evaluation device 5, and transmits (outputs) it to the evaluation device 5. For example, the behavior information extraction unit 49 extracts behavior information within a specific period from the behavior history table T2 in which the behavior history information is recorded, and further compiles the information for each person based on the person's identification information and transmits it to the evaluation device 5.

[0076] FIG. 5 shows an example of behavioral history information extracted by the behavioral information extraction unit 49 and transmitted to the evaluation device 5 when the behavioral history information of a person on January 24, 2023 is requested from the evaluation device 5. FIG. 5 shows an example in which a tenant name or a station name that is a destination facility is recorded as a point where a person was confirmed in the behavioral history information. However, instead of the tenant name or station name that is the destination facility, for example, identification information indicating the tenant or station that is the destination facility may be recorded, and information indicating the attributes of the tenant or station that is the destination facility may be recorded.

[0077] 2, the evaluation device 5 may be configured by a computer equipped with known hardware such as a processor 51 (CPU, MPU, etc.), a memory 53 (RAM, ROM, etc.), a storage 55 (a storage device such as an HDD or SSD), and a communication interface 57. The evaluation device 5 may also include a display and an input device (a keyboard, a mouse, etc.).

[0078] As shown in FIG. 3, the evaluation device 5 has, as functional units, a memory unit 61 that stores various information required for evaluation, a communication unit 63 for communicating with external devices, a clustering unit 65 that classifies people (clustering), and an evaluation execution unit 67 that executes evaluation based on the clustering results of the clustering unit 65.

[0079] The memory unit 61 is configured by the storage 55, and the clustering unit 65 and the evaluation execution unit 67 can be configured by the processor 51 executing a predetermined program. The communication unit 63 includes a communication interface 57. The communication unit 63 performs wireless or wired communication with other external devices (such as the behavior history server 13 and the terminal 7) via the network 21 in accordance with a known communication protocol.

[0080] The clustering unit 65 performs clustering based on the behavior history information acquired by the behavior history acquisition device 3, classifies people with similar characteristics, and generates classes (groups).

[0081] In the present invention, the clustering unit 65 performs clustering based on the behavior history information of the person, using the sum of the costs incurred by each person when traveling to a destination (in this embodiment, a destination facility) within a predetermined period of time.

[0082] However, the cost here indicates the price for travel to the destination, and is set based on at least one of the following: the distance traveled by the person to reach the destination, the travel time to reach the destination, and the cost (transportation cost) required for travel to reach the destination, the number of stations between the station where the person boarded the bus and the station nearest to the destination (store, etc.) (the number of sections between stations, the number of stations between stations, etc.), the distance between stations, the number of stops from boarding the bus to getting off, or the distance from boarding to getting off. When using a bus, it is preferable to be configured to be able to obtain position confirmation information for the boarding point and the alighting point by acquiring the boarding point and the alighting point at a touch point (e.g., a settlement device) installed in the bus. Similarly, in a taxi, it is preferable to be configured to be able to obtain position confirmation information for the boarding point and the alighting point by acquiring the boarding point and the alighting point at a touch point. In addition, the cost may be set based on information on whether the person is within or outside the range of a commuter pass owned by the person. The cost may be any of the parameters indicating the price for travel, or a combination of multiple parameters. In this way, by using at least one of the travel distance, travel time, cost (transportation cost), number of stations, distance between boarding point and disembarking point, etc., the cost as compensation for travel can be obtained appropriately and simply.

[0083] The greater the cost spent to reach a destination, the greater the person's interest in the destination area (the higher the interest level). For example, a class of people who spend a lot of money traveling to a theme park is inferred to be a class of people who show a strong interest in that theme park.

[0084] Therefore, by performing clustering based on the cost of travel to a destination, it is possible to extract a class of people with similar characteristics who are interested in a destination (e.g., a theme park). By analyzing the behavior of the extracted class, it is possible to grasp the average behavior of people belonging to that class and predict their behavior. This makes it possible to carry out effective advertising by, for example, extracting locations where people belonging to that class are likely to stop by and delivering advertisements to those locations that are predicted to interest people belonging to that class.

[0085] The evaluation execution unit 67 acquires from the terminal 7 a number of areas (hereinafter, candidate areas) that are candidates for the relocation of the business operator's facility. After that, the evaluation execution unit 67 evaluates the candidate areas by calculating an evaluation value (also called a score value) for each of the candidate areas using the clustering results. A candidate area with a high frequency of use by people belonging to a class that has a strong interest in the business operator's facility (business facility) is considered to be more suitable for location than an area with a low frequency. Therefore, the evaluation execution unit 67 calculates, as the evaluation value, a representative value of the frequency of use of the candidate area for each clustered class, weighted by a representative value of the degree of interest in the facility (business facility) operated by the business operator for each class, and summed up for all classes. In this way, the evaluation value is calculated by weighting and summing the frequency of use of the candidate area according to the degree of interest in the business facility of the business operator for each class, so that the evaluation value can be acquired so as to better reflect the degree of use expected when the business operator is located in the candidate area. When the calculation of the evaluation value is completed, the evaluation execution unit 67 outputs the calculated evaluation value to the terminal 7.

[0086] When calculating the evaluation value, the evaluation execution unit 67 uses the statistical value of the number of times stations are used in the candidate areas of people belonging to the class as a representative value of the number of times the candidate area is used for each class. In this case, the evaluation execution unit 67 uses the average number of times of use as the statistical value of the number of times of use. However, the present invention is not limited to this aspect, and the evaluation execution unit 67 may use the median or mode of the number of times of use as the statistical value of the number of times of use.

[0087] When calculating the evaluation value, the evaluation execution unit 67 uses a statistical value of the cost spent by people belonging to a class in traveling to a business facility or a destination facility similar to a business facility during a specified period as a representative value of the degree of interest in the business facility for each class. In this case, the evaluation execution unit 67 uses an average cost value as the cost statistical value. However, the present invention is not limited to this aspect, and the evaluation execution unit 67 may use a median or mode of the cost as the cost statistical value.

[0088] Next, the clustering unit 65 will be described in detail. The clustering unit 65 includes a behavior history request unit 71, a cost calculation unit 73, an attribute vector generation unit 75, a clustering execution unit 77, and a statistics generation unit 79. The behavior history request unit 71, the cost calculation unit 73, the attribute vector generation unit 75, the clustering execution unit 77, and the statistics generation unit 79 are each configured by the processor 51 of the evaluation device 5 executing a predetermined program.

[0089] The behavior history request unit 71 requests behavior history information from the behavior history server 13 and acquires the behavior history information. In this embodiment, the behavior history request unit 71 acquires behavior history information from the behavior history server 13 from the present to a predetermined period (hereinafter, analysis period, for example, several months to one year) ago as a behavior history log. FIG. 5 shows one day's worth of behavior history information from the behavior history log.

[0090] The cost calculation unit 73 acquires the destination (i.e., the destination facility) set according to the business operator, and acquires the sum of the costs (hereinafter, the total cost) spent by each person to reach each destination facility within the analysis period from the behavior history information acquired by the behavior history request unit 71.

[0091] 6 and 7 show examples of a process in which the cost calculation unit 73 calculates the total cost for each person and each destination facility. Here, it is assumed that tenants a, b, c, etc. are each set as a destination facility in advance.

[0092] The cost calculation unit 73 extracts destination facilities from one day's worth of behavior history information, as shown in Fig. 6. In Fig. 6, the extracted destination facilities (tenants a, b, c, etc.) are shown surrounded by a two-dot chain line rectangle. When the destination facility is found, the cost calculation unit 73 estimates the departure point by tracing back the behavior history, as shown in Fig. 7, and acquiring the position of the person before the time of arrival at the destination facility (the colored area in Fig. 7). By estimating the departure point in this way, the cost spent to reach the destination (destination facility) from the departure point can be easily and appropriately acquired.

[0093] Regarding the estimation of the departure point, if a person has stayed at the same place for a certain period of time at the time of arrival at the destination facility, the cost calculation unit 73 may set the place where the person stayed as the departure point. For example, in the examples of Fig. 6 and Fig. 7, since person A stayed at company X from 8:30 to 18:30, company X is estimated to be the departure point, and company X is set as the departure point. For person C, since his / her behavior history is not acquired until 9:00, it is estimated that he / she stayed at the same place for a certain period of time, and station S15 is set as the departure point.

[0094] Next, the cost calculation unit 73 obtains the cost that each person spent from the departure point to the destination facility (destination) on each day. After that, the cost calculation unit 73 adds up the costs that each person spent during the analysis period to obtain the total cost for each person and each destination facility.

[0095] 8(A) shows a table (hereinafter, attribute table T3) indicating the total cost for each destination facility for each person during the analysis period calculated by the cost calculation unit 73. Here, the analysis period is set to one year (from January 24, 2022 to January 23, 2023), and the number of stations passed on the way to the destination (number of distances between stations) is used as the cost.

[0096] FIG. 8(A) shows the level of interest of each person in each destination facility, and the level of interest is indicated by the total cost spent by each person at each destination facility within the analysis period. For example, the first line of FIG. 8(A) shows, horizontally, the total cost spent by person A to reach the destination facility within the analysis period. That is, each line of attribute table T3 shows, horizontally, the cost required for each person to reach each destination facility (tenants a, b, c, etc.) within the analysis period. That is, each line of attribute table T3 represents the level of interest of the person corresponding to that line for each destination facility. Below, the total number of people whose behavioral history was obtained during the analysis period and whose movement to the destination facility was confirmed is denoted by N. u It is written as follows.

[0097] The total cost spent by each person at each destination facility during the analysis period can also be calculated by multiplying the number of times the person visited each destination facility during the analysis period by the average cost spent to reach the destination facility during the analysis period. For example, if it is difficult to imagine that the person will travel to the destination facility from a place other than his / her home, the cost calculation unit 73 may generate the attribute table T3 by multiplying the cost required to travel from the nearest station to the home to the nearest station to the destination facility by the number of times the destination facility was used during the analysis period.

[0098] 8B shows a table indicating the average cost (average cost, here, the average distance between stations) that the person spent to reach each destination facility during the analysis period and the number of visits to the destination facility (number of visits). The cost calculation unit 73 may use this table to generate the attribute table T3 shown in FIG. 8A.

[0099] The attribute vector generation unit 75 generates attribute vectors for clustering (classifying / categorizing) people whose behavioral histories have been acquired.

[0100] The attribute vector is a vector indicating the attributes of the clustering target. The attribute vector generation unit 75 generates a vector indicating the attributes of each person (person vector) as an attribute vector in order to classify the people whose behavioral history has been acquired and whose total cost has been calculated.

[0101] The attribute here means the characteristics such as the preferences of each person, and in the present invention, the attribute represents the degree of interest of the corresponding person in each destination facility, that is, the interest degree (also called interest level). The person vector corresponds to a vector whose elements are the interest degree of the corresponding person in each destination facility (tenant a, b, c, etc.).

[0102] In this embodiment, the attribute vector generation unit 75 obtains the degree of interest of a person in each destination facility by the sum of the costs spent by the person in traveling to the destination facility during a predetermined period (e.g., an analysis period). Therefore, the attribute vector generation unit 75 generates a person vector as a vector whose element is the sum of the costs spent by the person corresponding to the person vector in traveling to the destination facility during a predetermined period (e.g., an analysis period).

[0103] That is, the attribute vector generation unit 75 generates a person vector U as an attribute vector based on the following formula (1): i (where i is the index indicating the person. i=1~N u )

[0104]

number

[0105] However, in formula (1), C ij represents the sum (total cost) of the costs incurred by person i to reach the destination facility j (tenant j) during the analysis period. N t indicates the total number of destination facilities (tenants).

[0106] The attribute vector generation unit 75 extracts each row of the attribute table T3 and generates a vector including the total costs as elements in order, thereby generating a person vector U based on the formula (1). i It is recommended to obtain the following.

[0107] In another embodiment, the attribute vector generation unit 75 generates a person vector as a vector having elements of the total cost spent by a person corresponding to the person vector in moving to a destination facility during a predetermined period (e.g., an analysis period) and the number of times each candidate area has been used. At this time, the attribute vector generation unit 75 generates a person vector U' as an attribute vector based on the following formula (1)'. i (where i is the index indicating the person. i=1~N u )

[0108]

number

[0109] In the formula (1), n ij represents the total number of times that person i used candidate area j during the analysis period. Note that N in formula (1) s represents the total number of candidate areas.

[0110] The attribute vector generation unit 75 calculates n ij is obtained based on the number of times that station Sj located in the candidate area Sj is used by person i. In this case, the attribute vector generation unit 75 may generate an attribute table T3 in which the total cost of each destination facility and the number of times that each station is used are arranged horizontally for each person shown in Fig. 8(C), and generate an attribute vector by extracting each row.

[0111] The clustering execution unit 77 performs classification of people, i.e., clustering, using the person vector generated by the attribute vector generation unit 75. The clustering performed by the clustering execution unit 77 may be based on a known method such as the k-means method. By the clustering execution unit 77 performing clustering, people whose behavioral history has been acquired are classified (i.e., classified).

[0112] Fig. 9 is an explanatory diagram of the clustering performed by the clustering execution unit 77. The clustering execution unit 77 uses the person vectors to classify people based on the degree of interest in each destination facility. In Fig. 9, the person vectors are each represented by a point on a graph, and an example of classification into three classes (classes α, β, and γ) is shown.

[0113] The statistics generating unit 79 obtains a representative value (summary value) indicating the characteristics of each class, such as a statistical average of the behaviors and characteristics of people belonging to the class and a distribution, based on the behavior information of people belonging to the class.

[0114] In the present invention, the statistics generating unit 79 refers to the clustering result and the behavior history information to obtain statistics on the degree of interest in the business facility to which the entrepreneur wishes to relocate.

[0115] Here, the statistics generating unit 79 calculates the sum of the costs spent by people belonging to each class to reach each destination facility (tenant) during the analysis period, and obtains the representative value of the sum as the statistic (statistical value) of the interest level of each destination facility for each class. Here, the statistics generating unit 79 uses the average value of the costs spent to reach each destination facility as the representative value of the costs spent to reach each destination facility. That is, the statistics generating unit 79 calculates the interest level of each destination facility for each class based on the following formula (2).

[0116]

number

[0117] In formula (2), w jk represents the interest level of class k for destination facility j. k represents the set of indices of people in class k, and N k represents the number of people in class k. Also, C ij represents the sum of the costs (total cost) spent by person i to reach tenant j during the analysis period. Therefore, equation (2) expresses the interest level of class k for the destination facility j, w jk is the total cost C ij It shows that the representative value of the interest level is represented by the average value of the total costs. By obtaining a representative value of the interest level by the representative value of the total costs in this way, the interest level of each class can be obtained easily and appropriately. Furthermore, by obtaining a representative value of the total costs by the average value of the total costs, the representative value of the total costs can be obtained easily and simply.

[0118] Here, the statistics generation unit 79 uses the average value of the costs spent to reach each destination facility as a representative value of the costs spent to reach each destination facility, but is not limited to this form, and various statistics (statistical values) that indicate the characteristics of a group, such as the median or mode, can be used instead of the average value.

[0119] The statistics generating unit 79 calculates the degree of interest w of class k for the destination facility j. jk When calculating, a person belonging to class k is extracted from the attribute table T3, and the total cost C ij 10(A) and 11(A) show the degree of interest w for the destination facility for each class calculated by the statistics generating unit 79. jk A portion of a table showing an example of the interest level is shown below (hereinafter, referred to as interest level table T4).

[0120] When a business facility is included in the destination facilities, the statistics generation unit 79 calculates the statistical value of the interest level of each class in the business facility based on the average of the total cost for people belonging to that class to reach the business facility, based on equation (2).

[0121] When a business facility is not included in the destination facilities, the statistics generation unit 79 may extract a destination facility similar to the business facility and obtain the statistical value of the interest level of each class for that destination as the statistical value of the interest level for the business facility.

[0122] Next, a detailed description will be given of the evaluation execution unit 67. The evaluation execution unit 67 includes an area information acquisition unit 81, an evaluation value calculation unit 83, and an optimum area output unit 85.

[0123] The area information acquisition unit 81 calculates a representative value of the number of times of use of candidate areas that are candidates for relocation destinations for each class generated by the clustering unit 65. In this embodiment, based on the input (selection) of the candidate areas of the business operator acquired by the terminal 7, the area information acquisition unit 81 calculates a representative value of the number of times of use of the candidate areas corresponding to each class by the average number of times (number of times of use) that people belonging to that class used the stations and stops (including bus stops, etc.) corresponding to the candidate areas during the analysis period.

[0124] That is, the area information acquisition unit 81 calculates a representative value of the number of times the candidate areas for each class are used based on the formula (3).

[0125]

number

[0126] In formula (3), M kl represents the representative value of the number of times the candidate area of ​​class k is used. Also, in equation (3), Ω k represents the set of indices of people in class k, and N k represents the number of people in class k. Also, m ilis the candidate area Sl (l = 1 to N p ) corresponding to station Sl (l=1~N p Equation (3) expresses the statistical average number of times class k is used at station Sl, M kl is the number of times that person i who belongs to class k uses station Sl (candidate area Sl) m il In this way, by using the average value of the number of uses, a representative value of the number of uses can be easily and appropriately obtained.

[0127] When calculating the average number of times of use of class k for station S1, the area information acquisition unit 81 refers to the behavior history information and calculates the number of times each person has used station S1 during the analysis period m as shown in FIG. il After that, the area information acquisition unit 81 may create a table (hereinafter, referred to as an individual use frequency table T5) of the number of times that a person j belonging to class k has used station S1. il By averaging these, the average number of times that people belonging to class k use station Sl is calculated as M kl FIG. 11B shows the average value M kl 13 shows a portion of a table (usage frequency table T6) indicating the above.

[0128] The evaluation value calculation unit 83 calculates an evaluation value when the business facility is moved to the candidate area based on the clustering result. Here, the evaluation value is calculated so that it becomes higher as the expected number of uses increases.

[0129] In this embodiment, the evaluation value calculation unit 83 weights the number of times the candidate area for each class is used (specifically, the number of times the corresponding station is used) by the degree of interest in the business facility for that class, and adds up the numbers for all classes to calculate the evaluation value when a facility (tenant) is moved to a candidate area.

[0130] When calculating the evaluation value, the evaluation value calculation unit 83 determines a destination facility associated with the business facility. When the business facility is included in the destination facilities (for example, when the business facility is relocated), the evaluation value calculation unit 83 sets the business facility itself as a destination facility associated with the business facility, and obtains an index indicating the business facility.

[0131] When a business facility is not included in the destination facilities (for example, when a new business facility is being established), the evaluation value calculation unit 83 determines that a destination facility similar to the business facility is a destination facility that can be associated with the business facility, and obtains an index indicating the destination facility similar to the business facility.

[0132] Specifically, when an operator intends to establish a new business facility, the evaluation value calculation unit 83 may extract, from among the destination facilities, destination facilities that handle products in the same genre as the products handled by the operator as destination facilities that can be associated with the business facility. When an operator intends to establish a new directly managed store or franchise store, the evaluation value calculation unit 83 may extract other directly managed stores or franchise stores operated by the operator as destination facilities that can be associated with the business facility.

[0133] Next, the evaluation value calculation unit 83 calculates an evaluation value of the candidate area when the business facility is established, based on the following formula (4).

[0134]

number

[0135] In formula (4), E al represents the evaluation value of the candidate area l (area surrounding station Sl) for the location of the business facility. In formula (4), a represents the index a of the destination facility (tenant) associated with the business facility. In formula (4), Σ means the sum of the product of the interest level (weight) and the number of times of use for each class for all classes.

[0136] That is, the evaluation value E when the business facility is relocated to the candidate area l alis expressed as (average total cost for destination facility a corresponding to business facilities of class α) × (average number of times of use of station Sl of class α) + (average total cost for destination facility a corresponding to business facilities of class β) × (average number of times of use of station Sl of class β) + ····

[0137] At this time, the evaluation value calculation unit 83 may calculate the evaluation value of each candidate area when the destination facility is installed in each candidate area, and generate an evaluation value table T7 that summarizes the evaluation values ​​in the form of a table. An example of the evaluation value table T7 is shown in Fig. 11(C).

[0138] In this embodiment, the evaluation value calculation unit 83 calculates E al (l=1~N p ) among the maximum value E max,a Extract E al E max,a Divide by and multiply by 100 to get the normalized normal evaluation value e al That is, the normal evaluation value e al is expressed by the following equation (5).

[0139]

number

[0140] FIG. 11D shows the normal evaluation value e calculated by the evaluation value calculation unit 83. al A table showing an example of this (hereinafter, referred to as a normal evaluation value table T8) is shown.

[0141] The optimum area output unit 85 acquires an evaluation value (or a normal evaluation value) of each candidate area for each destination facility, and extracts the candidate area with the highest evaluation value as the optimum candidate area. After that, the optimum area output unit 85 outputs information (identification information, index, etc.) indicating the optimum candidate area to the terminal 7.

[0142] In this embodiment, the optimum area output unit 85 outputs the normal evaluation value of each candidate area to the terminal 7 together with information indicating the optimum candidate area.

[0143] When a business facility is associated with a destination facility a, the optimal area output unit 85 outputs the candidate area S2 (station S2) with the highest normal evaluation value from the row corresponding to the destination facility a in the normal evaluation value table T8 of Figure 11 (D) as the optimal candidate area.

[0144] Similarly, the optimum area output unit 85 outputs candidate area S1 as the optimum candidate area when the business facility is associated with destination facility b, and outputs candidate area S2 as the optimum candidate area when the business facility is associated with destination facility c.

[0145] The terminal 7 may be configured by a computer equipped with known hardware such as a processor 91 (CPU, MPU, etc.), memory 93 (RAM, ROM, etc.), storage 95 (storage device such as HDD or SSD), communication interface 97, display device 99 (display, touch panel display), and input device 101 (keyboard, mouse, touch panel display, etc.).

[0146] The terminal 7 includes, as functional units, a communication unit 103, a candidate area acquisition unit 105, and an evaluation value display unit 107.

[0147] The communication unit 103 includes a communication interface 97. The communication unit 103 performs wireless or wired communication with other external devices (such as the behavior history server 13 and the evaluation device 5) via the network 21 in accordance with a known communication protocol.

[0148] The candidate area acquisition unit 105 can be configured by the processor 91 of the terminal 7 executing a predetermined program, acquiring an input at the input device 101 , and controlling the display device 99 and the communication interface 97 .

[0149] The candidate area acquisition unit 105 displays a candidate area input screen 201 on the display device 99 (display), acquires an input related to the candidate area at the input device 101, and outputs the selected candidate area to the evaluation device 5.

[0150] FIG. 12 shows an example of a candidate area input screen 201 .

[0151] In the candidate area input screen 201, check boxes 203 for selecting candidate areas are superimposed on a map 202 showing an area where a business operator is considering establishing a facility. The check boxes 203 are preferably arranged to correspond to stations. When a user of the evaluation system 1 checks the check boxes 203 and presses an evaluation start button 204, the candidate area acquisition unit 105 transmits (outputs) information related to the selected candidate area to the evaluation device 5.

[0152] The evaluation value display unit 107 can be configured by the processor 91 of the terminal 7 executing a predetermined program and controlling the display device 99 .

[0153] After the candidate area acquisition unit 105 outputs the selected candidate area to the evaluation device 5, the evaluation value display unit 107 displays a standby screen 210 on the display device 99 until a regular evaluation value is acquired from the evaluation device 5. An example of the standby screen 210 is shown in Fig. 13. The standby screen 210 may include a progress bar 211 that indicates the progress of the processing by the evaluation device 5.

[0154] When the evaluation value display unit 107 acquires from the evaluation device 5 the candidate areas most suitable as relocation destinations for the facilities of a business operator considering relocation and the normal evaluation values ​​for each of the candidate areas, it causes a result display screen 220 including these to be displayed on the display device 99. However, the result display screen 220 may display the evaluation values ​​instead of the normal evaluation values, or may display both the evaluation values ​​and the normal evaluation values.

[0155] 14 shows an example of a result display screen 220. The result display screen 220 displays a map 202 showing the area around the facility of the business operator, with a popup 221 for displaying the normal evaluation value and the optimal candidate area superimposed thereon.

[0156] Next, the operation of the evaluation system 1 configured in this manner will be described. The evaluation system 1 is used by a business operator who is considering relocating their facility. The business operator's facilities are assumed to be predetermined, and the facilities to be used for clustering are assumed to be input in advance by the operator of the evaluation system 1. In addition, it is assumed that the behavioral history of people is acquired using the imaging device 11 in all facilities, including the business operator's facilities.

[0157] When an input indicating the start of evaluation is made to the evaluation device 5, the evaluation device 5 executes the evaluation program, performs evaluation processing for implementing the evaluation method, and evaluates candidate areas for the location of the business facility. Figure 15 shows a sequence diagram of the evaluation processing.

[0158] As shown in FIG. 15, when the evaluation program is started, the evaluation device 5 transmits a request to start accepting input of candidate areas (ST101).

[0159] When the terminal 7 receives a request to start accepting input of candidate areas from the evaluation device 5, it displays the candidate area input screen 201 shown in Fig. 12 and accepts the input of the candidate areas (ST102). When the acceptance is completed, the terminal 7 transmits the acquired information related to the candidate areas to the evaluation device 5 (ST103).

[0160] When the evaluation device 5 receives the information related to the candidate area from the terminal 7, it requests the behavior history server 13 to transmit behavior history information for the analysis period (ST104).

[0161] When the behavior history server 13 receives a request to transmit behavior history information from the evaluation device 5, it extracts behavior information for the analysis period and transmits it to the evaluation device 5 (ST105).

[0162] When the processor 51 of the evaluation device 5 acquires the behavior information for the analysis period, it extracts, for each person, movements during the analysis period with each destination facility as the destination, and calculates the cost to reach the destination facility for each movement. The processor 51 of the evaluation device 5 then adds up the costs for each destination facility to acquire the sum of the costs (total cost) during the analysis period for each person by adding up the costs for each destination facility. As a result, the processor 51 of the evaluation device 5 generates an attribute table T3 with each total cost as an element (ST106).

[0163] The cost (total cost) that a person spent on traveling to each of the destination facilities during the analysis period and the related indexes are considered to represent the person's degree of interest (also called interest level) in each of the facilities. Therefore, the processor 51 of the evaluation device 5 can be said to obtain the person's degree of interest in the facilities by calculating the total cost of each facility and the related indexes.

[0164] When the generation of the attribute table T3 is completed, the processor 51 of the evaluation device 5 generates a person vector indicating the attributes of each person, using the total cost included in the attribute table T3. After that, the evaluation device 5 executes clustering to classify people using the generated person vector (ST107). The generation of the attribute table (ST106) and the clustering by the attribute vector (ST107) constitute an analysis step to classify people based on the behavior information of people moving in a public space.

[0165] When the clustering is completed, the processor 51 of the evaluation device 5 generates an interest level table T4 (ST108). As shown in Fig. 11(A), the interest level table T4 is a table in which a representative value of the interest level in facilities (tenants) including business facilities of each class is recorded, and as the representative value of the interest level, a statistical value (in this embodiment, an average value) of the cost spent by a person belonging to the class until he or she reaches a facility corresponding to the business facility (business facility or a destination facility corresponding to the business facility) during a predetermined analysis period is used.

[0166] When the interest level table T4 is acquired, the processor 51 of the evaluation device 5 generates a use count table T6 (ST109). As shown in FIG. 11(B), the use count table T6 is a table in which a representative value of the use count in the analysis period of the candidate area for each class is recorded, and the representative value of the use count is the number of times that the candidate area is used by the person belonging to the class, that is, the statistical value of the use count (average value in this embodiment). Therefore, the processor 51 of the evaluation device 5 calculates the average value of the use count of the stations located in the candidate area of ​​the person belonging to the class as the representative value of the use count in the analysis period of the candidate area, and generates the use count table T6.

[0167] When the generation of the use count table T6 is completed, the processor 51 of the evaluation device 5 refers to the interest level table T4 and the use count table T6 for each candidate area, and calculates an evaluation value by adding up the product of the representative value of the interest level in the business facility and the representative value of the use count of the candidate area for each class. In other words, the evaluation value corresponds to the sum of the representative value of the use count of the candidate area for each class, weighted by the average value of the use count of the candidate area.

[0168] At this time, the processor 51 of the evaluation device 5 may calculate an evaluation value when each of the destination facilities is set up in each candidate area, and generate an evaluation value table T7. When a business facility is included in the destination facilities, the processor 51 of the evaluation device 5 may refer to the interest level table T4, obtain a representative value of the interest level corresponding to the business facility, and calculate an evaluation value. When a business facility is not included in the destination facilities, such as when a new business facility is set up, the processor 51 of the evaluation device 5 may extract a destination facility similar to the business facility, refer to the interest level table T4, obtain a representative value of the interest level corresponding to the extracted destination facility, and calculate an evaluation value.

[0169] When the calculation of the evaluation value is completed, the processor 51 of the evaluation device 5 calculates a regular evaluation value by normalizing based on the calculated evaluation value, and extracts the most suitable candidate area for installation (ST110). The generation of the interest level table (ST108), the generation of the number of times of use table (ST109), and the extraction of the evaluation value, the regular evaluation value, and the most suitable candidate area for location constitute an evaluation step of evaluating the candidate areas based on the results of the analysis step.

[0170] When the calculation of the normal evaluation value and the extraction of the optimal candidate area for the location are completed, the processor 51 of the evaluation device 5 transmits each of the normal evaluation value and information related to the optimal candidate area for installation to the terminal 7 (ST111).

[0171] When the terminal 7 acquires the information related to the normal evaluation value and the optimal candidate area for the location, it displays the result display screen 220 shown in Fig. 14 (ST112). By displaying the result display screen 220, the user of the evaluation system 1 can understand the normal evaluation value of each candidate area and the optimal candidate area for the location.

[0172] Next, the effects of the evaluation system 1 configured as above will be described. For simplicity, the explanation will be given by taking as an example a case where evaluation is performed based on the behavioral histories of four persons (person A, person B, person C, and person D).

[0173] Here, it is assumed that person A and person B each belong to the F1 class (women aged 20 to 34), and person C and person D each belong to the F3 class (women aged 50 or older).

[0174] Tenant a and b are both popular stores among the F1 demographic, Tenant a is a fashion store, Tenant b is a cafe that offers sweets, and Tenant c and b are both stores that sell goods that are popular among the F3 demographic.

[0175] 16(A) shows a table (attribute table T3) of the total costs spent by persons A to D on traveling to tenants a to d during the analysis period. Here, the analysis period is set to three months, and the type of cost used is not the cost to reach the destination, but the travel time required to reach the destination.

[0176] As shown in FIG. 16(A), both persons A and B are interested in the products for the F1 demographic handled by tenants a and b and use them frequently, so the cost spent on travel to tenants a and b is also higher than that spent on travel to tenants c and d. Therefore, it can be said that persons A and B are more interested in tenants a and b than tenants c and d, and as described above, it can be said that the attribute table T3 represents the interest level of each of persons A to D in each of tenants a to d. On the other hand, both persons C and D are interested in the products for the F3 demographic handled by tenants a and b and use them frequently, so the cost spent on travel to tenants c and d is also higher than that spent on travel to tenants a and b.

[0177] Tenant a is located near station S1, tenant b near station S2, and tenants c and d near station S3. The areas around station S1 and station S2 are both famous for having fashionable stores that are popular with the F1 demographic. The area around station S1 in particular has recently become popular with the F1 demographic, and the number of visitors has increased sharply. The area around station S3 is famous as a town that is popular with the F3 demographic.

[0178] Figure 16(B) shows a table (individual use count table T5) of the number of times that people A to D used stations S1 to S3 during the analysis period. People A and B have recently been visiting the areas around stations S1 and S2, which are popular with the F1 demographic, but they visit station S1 more frequently than station S2. People C and D often use the area around station S3, which is popular with the F3 demographic.

[0179] Here, assume a situation in which a business operator managing tenant b inputs stations S1 to S3 as candidate areas into the evaluation system 1 in order to consider an area to which tenant b should relocate.

[0180] 17 shows a graph in which the horizontal axis represents the total cost spent on traveling to tenant a and the vertical axis represents the total cost spent on traveling to tenant c. On the graph, points corresponding to person vectors corresponding to persons A to D generated by the evaluation system 1 are shown. As can be seen from FIG. 17, persons A to D are classified into class α consisting of persons A and B and class β consisting of persons C and D according to the type of tenant they are interested in.

[0181] 18(A) shows the interest level table T4 generated by the evaluation system 1 at this time. Since class α includes many people belonging to the F1 tier, the interest level in tenants a and b, which are popular among the F1 tier, is higher than that in tenants c and d. Similarly, since class β includes many people belonging to the F3 tier, the interest level in tenants c and d, which are popular among the F3 tier, is higher than that in tenants a and b.

[0182] 18(B) shows the frequency of use table T6 generated by the evaluation system 1 at this time. Since class α includes many people belonging to the F1 demographic, the frequency of use of stations S1 and S2, which are located in towns popular with the F1 demographic, is higher than that of station S3. Similarly, since class β includes many people belonging to the F3 demographic, the frequency of use of station S3, which is located in towns popular with the F3 demographic, is higher than that of stations S1 and S2.

[0183] Figures 18 (C) and (D) respectively show a table (evaluation value table T7) showing a list of evaluation values ​​generated by the evaluation system 1 at this time, and a table (normal evaluation value table T8) showing a list of normal evaluation values.

[0184] When the operator of tenant b inputs stations S1 to S3 as candidate areas, the evaluation system 1 outputs the area around station S1, which has the highest normal evaluation value, as the most suitable candidate area for location, as can be seen from Figures 18 (C) and (D).

[0185] Tenant b, operated by a business operator, is targeted at the F1 demographic, and is therefore of high interest to people belonging to class α. Therefore, by installing Tenant b in station S1, which is frequently used by class α, it is expected that the use of Tenant b will be promoted. Therefore, it can be understood that the evaluation system 1 outputs candidate areas suitable for the location of Tenant b.

[0186] In this way, the degree of interest in a business facility can be appropriately obtained by obtaining the degree of interest from the average value of the total cost for travel to the destination facility set for that business facility during the analysis period. In addition, by using the degree of interest and the number of times the candidate area is used, an evaluation value can be obtained that corresponds to the expected degree of use when a business operator sets up in the candidate area. This allows the evaluation system 1 to present candidate areas suitable for a location to the business operator.

[0187] As described above, the embodiments have been described as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited to these, and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are made. In addition, it is also possible to combine the components described in the above embodiments to create new embodiments.

[0188] In the above embodiment, in the evaluation system 1, the clustering unit 65 performs clustering using the cost, and the evaluation execution unit 67 calculates the evaluation value based on the cost, but the present invention is not limited to this embodiment. The evaluation system 1 may be configured such that the clustering unit 65 performs clustering without using the cost, and then the evaluation execution unit 67 performs evaluation based on the cost, as in the above embodiment. The evaluation system 1 may be configured such that the clustering unit 65 performs clustering using the cost, and then the evaluation execution unit 67 performs evaluation based on the clustering result, without using the cost. In addition, the evaluation system 1 is configured to display the normal evaluation value and the optimal candidate area for installation on the result display screen 220, but the evaluation system 1 may be configured to display the evaluation value (or the optimal candidate area) on the result display screen 220.

[0189] In the above embodiment, the evaluation device 5 is provided with the clustering unit 65 and the evaluation execution unit 67, but they may be configured as separate devices (clustering device and evaluation value calculation device). In this case, the clustering device does not transmit behavioral information of each person included in the class to the evaluation value calculation device, and the information provided is the averaged numerical value (cost and number of uses) for each class. This makes it easier to protect personal information.

[0190] In this case, the analysis device is an analysis device for extracting users of facilities included in the public space, and is equipped with a behavioral information acquisition unit 47 that acquires behavioral information of people moving in the public space, and a clustering unit 65 that classifies users based on the behavioral information, and the clustering unit 65 can be configured to acquire, for each person, a total of costs that the person spent on using each facility within a specified period based on the behavioral information, and to classify the people by clustering the people based on the total costs.

[0191] In the above embodiment, the evaluation value was calculated by multiplying the average value of the total cost of each class by the average value of the number of times each class is used and adding them up, but any method may be used to weight the average value of the total cost of each class. For example, the evaluation value may be calculated by calculating a coefficient by substituting the average value of the total cost of each class into a function that depends on the average value of the total cost of each class, and adding up the product of the coefficient and the average value of the number of times each class is used. In addition, when the function includes multiple parameters, it is also possible to configure the parameters to be changed by machine learning.

[0192] In the above embodiment, the evaluation method, evaluation system, and evaluation program are described as being used to evaluate a relocation destination, but the present invention is not limited to this form and may also be applied to existing evaluations (areas where a facility is already located).

[0193] In the above embodiment, the interest level is calculated based on the sum of the costs each person spends to reach each of the destination facilities (facilities of use) within a predetermined analysis period, but the invention is not limited to this embodiment.

[0194] For example, the interest level may be obtained by multiplying the total cost that each person spent to reach each destination (destination facility) during the analysis period by the number of times that the person visited the destination facility during the analysis period (i.e., the number of visits). Figure 19 shows an example of an attribute table T3 corresponding to Figure 8(B) when the interest level is represented by the product of the total cost and the number of visits.

[0195] In this way, the level of interest is not limited to the sum of the costs each person spends to reach each destination facility (facility of use) within a specified analysis period, but may be anything based on the cost of traveling to each destination (destination facility) within a specified period and the number of visits to the destination (destination facility). [Industrial Applicability]

[0196] The evaluation method, evaluation system, and evaluation program disclosed herein are useful as an evaluation method, evaluation system, and evaluation program for conducting an appropriate evaluation to select a candidate area suitable for location from among multiple candidate areas included in a public space. [Explanation of symbols]

[0197] 1: Rating system 3: Behavioral history acquisition device 5: Evaluation device 7: Terminal 11: Imaging device 11A: Camera 11B: Camera 11C: Camera 11D: Camera 13: Action history server 15: Entrance / Exit Gate 19: Gate management device 21: Network 23: Authentication machine 31: Processor 33: Memory 35: Storage 37: Communication interface 41: Storage section 43: Communications Department 45:Person identification department 47: Behavioral information acquisition unit 49: Behavioral information extraction unit 51: Processor 53: Memory 55: Storage 57: Communication interface 61: Storage section 63: Communications Department 65: Clustering Department 67: Evaluation Execution Department 71: Action history request section 73: Cost calculation section 75: Attribute vector generation unit 77: Clustering execution unit 79: Statistics generation part 81: Area information acquisition unit 83: Evaluation value calculation unit 85: Optimal area output section 91: Processor 93: Memory 95: Storage 97: Communication interface 99:Display device 101: Input device 103: Communications Department 105: Candidate area acquisition section 107: Evaluation value display section 201: Candidate area input screen 202: Map 203: Checkbox 204: Evaluation start button 210: Standby screen 211: Progress bar 220:Result display screen 221: Pop-up

Claims

1. A method for evaluating candidate areas for a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, the method comprising the steps of: A processor performing the evaluation method, Extracting movement histories to destinations set according to the business operator from the behavioral history of each person moving through the public space during a predetermined period of time; Using the travel history, for each person, obtain a degree of interest in each of the destinations based on a cost related to travel to the destinations within a predetermined period and a number of visits to the destinations; acquiring a number of times each of the candidate areas has been used within the predetermined period for each of the persons; An evaluation method for obtaining an evaluation value for each of the candidate areas based on the degree of interest and the number of times of use.

2. The evaluation method according to claim 1 , wherein the cost is set based on at least one of a travel distance, a travel time, and an expense required to reach the destination.

3. The evaluation method according to claim 2 , wherein the processor extracts the destination from the behavior history, estimates a departure point by obtaining a position of the person before a time of arrival at the destination, and obtains the cost.

4. The processor performs clustering to classify the persons into classes using the sum of the costs for each of the destinations; The evaluation method according to any one of claims 1 to 3, wherein the evaluation value is obtained by adding up the product of a representative value of the sum of the costs for each of the classes and a representative value of the number of uses for each of the classes.

5. The evaluation method according to claim 4 , wherein the processor obtains the evaluation value by adding up the product of an average value of the sum of the costs for each of the classes and an average value of the number of times of use for each of the classes.

6. A method for evaluating candidate areas for a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, the method comprising the steps of: A processor performing the evaluation method, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; The processor, In the analyzing step, a degree of interest in each destination set according to the business operator is acquired for each person; For each of the persons, a vector having an interest level for each of the destinations as an element is obtained as an attribute vector indicating an attribute related to the interests of the person; An evaluation method for clustering based on the attribute vector.

7. A method for evaluating candidate areas for a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, the method comprising the steps of: A processor performing the evaluation method, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; In the evaluation step, the processor obtains a representative value of the number of times of use for each of the candidate areas for each class obtained by the analysis step and a representative value of the degree of interest in the business facility in which the operator intends to locate for each of the classes, and performs evaluation by weighting the representative values ​​of the number of times of use based on the representative value of the degree of interest and adding them together to obtain an evaluation value.

8. A candidate area evaluation system for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, comprising: a processor for performing the evaluation; The processor, Extracting movement histories to destinations set according to the business operator from the behavioral history of each person moving in the public space; Using the movement history, for each of the persons, an interest level for each of the destinations is obtained based on a cost related to movement to the destinations within a predetermined period and a number of visits to the destinations; acquiring a number of times each of the candidate areas has been used within the predetermined period for each of the persons; An evaluation system that obtains an evaluation value for each of the candidate areas based on the degree of interest and the number of times of use.

9. The processor performs clustering to classify the persons into classes using the sum of the costs for each of the destinations; The evaluation system according to claim 8 , wherein the evaluation value is obtained by adding up a product of a representative value of the sum of the costs for each of the classes and a representative value of the number of times of use for each of the classes.

10. A candidate area evaluation system for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, comprising: a processor for performing the evaluation; The processor, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; In the analyzing step, a degree of interest in each destination set according to the business operator is acquired for each of the persons; For each of the persons, a vector having an interest level for each of the destinations as an element is obtained as an attribute vector indicating an attribute related to the interests of the person; A rating system that clusters based on the attribute vectors.

11. A candidate area evaluation system for allowing a business operator to select a candidate area suitable for location from among a plurality of candidate areas included in a public space, comprising: a processor for performing the evaluation; The processor, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; In the evaluation step, the processor obtains a representative value of the number of times of use for each of the candidate areas for each class obtained by the analysis step, and a representative value of the degree of interest in the business facility in which the operator for each of the classes wishes to locate, and performs an evaluation by weighting and adding up the representative values ​​of the number of times of use based on the representative value of the degree of interest and obtaining an evaluation value.

12. A program for evaluating candidate areas, which is used by a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, a processor for performing the evaluation, Extracting movement histories to destinations set according to the business operator from the behavioral history of each person moving in the public space; Using the movement history, for each of the persons, an interest level for each of the destinations is obtained based on a cost related to movement to the destinations within a predetermined period and a number of visits to the destinations; acquiring a number of times each of the candidate areas has been used within the predetermined period for each of the persons; an evaluation program for obtaining an evaluation value for each of the candidate areas based on the degree of interest and the number of times of use;

13. The processor performs clustering to classify the persons into classes using the sum of the costs for each of the destinations; 13. The evaluation program according to claim 12, wherein the evaluation value is obtained by adding up a product of a representative value of the sum of the costs for each of the classes and a representative value of the number of times of use for each of the classes.

14. A program for evaluating candidate areas, which is used by a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, a processor for performing the evaluation, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; In the analyzing step, a degree of interest in each destination set according to the business operator is acquired for each of the persons; For each of the persons, a vector having an interest level for each of the destinations as an element is obtained as an attribute vector indicating an attribute related to the interests of the person; An evaluation program that performs clustering based on the attribute vectors.

15. A program for evaluating candidate areas, which is used by a business operator to select a candidate area suitable for a location from among a plurality of candidate areas included in a public space, a processor for performing the evaluation, An analysis step of clustering the people based on behavioral information of the people moving in the public space; an evaluation step of evaluating the candidate areas based on the results of the analysis step; The processor, in the evaluation step, obtains a representative value of the number of times of use for each of the candidate areas for each class obtained by the analysis step and a representative value of the degree of interest in the business facility in which the operator for each of the classes wishes to locate, and performs an evaluation by weighting and adding up the representative values ​​of the number of times of use based on the representative value of the degree of interest to obtain an evaluation value.

Images