JP6027122B2 - Information analysis system and information analysis method - Google Patents

Description

  The present invention relates to an information analysis system and an information analysis method, and, for example, relates to a technique for extracting a power range of an area using a degree of association between areas obtained from human movement data.

(Area analysis needs)
Understanding the “area characteristics” in marketing, etc. is important for corporate strategy. The analysis of area characteristics has the following needs.

(I) I want to know what purpose (work, return home, shopping, etc.) people are coming to the area.

(Ii) I want to know how far this area attracts people.

  For example, in the service industry, when a store targeting a person who goes shopping on a holiday along a railway line is developed, it is necessary to grasp in the store development strategy which station along the line has many shoppers. In addition, if there is a station that already has a store along the line, it is also important information on how far the station is currently attracting customers. This is because there is a possibility that customers of existing stores may be dispersed when there are many customers from stations that are newly opening stores.

(Conventional technology for area analysis needs)
Conventionally, the following techniques have been provided for such needs. That is, (a) a questionnaire survey, (b) an analysis using human movement data, and (c) a trade area analysis using a model. Details will be described below.

(Conventional technology for area analysis needs: questionnaire survey)
It is difficult to estimate / acquire information corresponding to the needs as described above from numerical values such as the number of passengers at a station and the population of a region.

  For this reason, in the questionnaire survey, information necessary for area characteristic analysis is collected by having a person who has come to a certain area directly answer what they came for. An example of a questionnaire survey is the “City Traffic Census” conducted by the Ministry of Land, Infrastructure, Transport and Tourism. For example, the metropolitan traffic census conducts a questionnaire survey for railway users. More specifically, questionnaires are distributed in each of the Tokyo, Chukyo, and Kinki areas, and the total number of railway users is less than 500,000. I have an answer. By using this result, it is possible to grasp where and for what purpose people are coming.

(Conventional technology for area analysis needs: Analysis using human movement data)
For example, Patent Literature 1 discloses a technology that uses a movement history of a mobile terminal as human movement data. In Patent Literature 1, the user's location information is acquired from a mobile terminal possessed by the user, and totaled and processed in real time, and the information is provided to the user. With this technology, it is possible to grasp how much people are moving from which area to which area.

  Further, Patent Document 2 discloses a technique for using a traffic IC card usage history as human movement data. In patent document 2, the movement range of the user who entered and exited a certain station using the utilization log | history of a traffic type | system | group IC card is made into the customer collection range of the said station.

(Conventional technology for area analysis needs: business area analysis using models)
For example, Non-Patent Document 1 discloses a trade area analysis using a Huff model as an example of a trade area analysis using a model. The Huff model is that the probability that a person living in a certain area will go shopping at a certain commercial facility increases as the sales area of the commercial facility increases, and conversely the greater the distance to the commercial facility. It is a model that becomes lower. With this model, when there are a plurality of commercial facilities, it is possible to obtain the probability of going to each commercial facility from a certain area (such as a town). By calculating the population of each town included in a wide area and the probability of shopping from each town to each commercial facility, it is possible to predict how many people gather from which town in each commercial facility.

JP 2006-1113837 A JP 2011-164811 A

David L. HUFF: "Defining and Estimating Trading Area", Journalof Marketing, Vol.28, pp34-38 (1964)

  By means described in Patent Documents 1 and 2 above, it is possible to realize the purpose of movement of people and the amount of movement, or the prediction of the number of customers attracted to the area. There is.

(Issue 1: Questionnaire survey)
Surveys by questionnaires have the merit of being able to perform analysis close to the actual situation because data is directly extracted from the person who is moving, but at the same time costs. Also, for example, in the metropolitan traffic census that the Ministry of Land, Infrastructure, Transport and Tourism conducts every five years, the collection rate of questionnaires for railway users is about 13-14%, which is not high. Furthermore, the aforementioned survey is for a day on weekdays, and it is difficult to obtain continuous questionnaire survey results in a wide area.

(Issue # 2: Analysis using human movement data)
The analysis using the movement data of people in Patent Documents 1 and 2 has an advantage that the data is automatically acquired from a mobile terminal carried by the user or a traffic IC card used at a station, so that there is no cost. . On the other hand, there is a problem that the purpose of the user's movement is unknown. As a result, it cannot meet the needs of area analysis that people want to know what they have moved for.

(Issue 3: Analyzing a trade area using a model)
A trade area analysis using a model has the advantage that it can easily predict the range of influence of a commercial facility if it can obtain the values of model parameters such as the sales floor area of the commercial facility and the distance to them, but it does not fit the model. Cases can also be seen. The Huff model mentioned above is often applied to the nearest daily items such as groceries, but it is often used for items that are not frequently purchased, such as brand clothing and furniture, with a focus on design and quality. May not match. For example, famous department stores and high-end brand street stores with high fashionability in the city center are visited by shoppers from the suburbs regardless of distance. Therefore, it is difficult to grasp the actual range from which a customer is attracted to a certain area using only a model based on the distance to a commercial facility.

  The present invention has been made in view of such a situation, and information on the purpose of visiting a predetermined place or region of people and the extent to which people are visiting a specific place or region is appropriately and We propose a technique to provide as cheaply as possible.

  In order to solve the above-described problem, in the present invention, in a computer system, a staying place of a person belonging to a person group is checked by comparing history data including a movement history of the person group with category data of a staying place prepared in advance. The stay extraction process for extracting the stay and the stay frequency extracted by the stay extraction process for each individual included in the human group, and for each individual for each stay purpose prepared in advance based on the stay frequency for each individual Based on base location information for each individual purpose of stay, frequency count processing for totaling the amount of movement between locations for each individual purpose of stay, Relevance calculation processing for calculating the relevance indicating how closely the places by stay purpose are related to each other based on the information on the amount of movement between places by stay purpose An information providing process for providing to information of relevance was to run.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to analyze at low cost by attracting customers to a place according to the purpose of staying at a moving destination as a power range.

  Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. The embodiments of the present invention can be achieved and realized by elements and combinations of various elements and the following detailed description and appended claims.

  It should be understood that the description herein is merely exemplary and is not intended to limit the scope of the claims or the application of the invention in any way.

It is a figure for demonstrating the association degree between the stations in this invention. 1 is a diagram showing a schematic configuration of an area analysis apparatus (area analysis system) 1 according to a first embodiment of the present invention. It is a figure which shows the hardware constitutions of the area analyzer 1 by the 1st Embodiment of this invention. It is a figure which shows the data structure of the IC card utilization log | history 103. It is a figure which shows the data structure of the stay table. It is a figure which shows the data structure of the station frequency table. It is a figure which shows the data structure of the base data. It is a figure which shows the data structure of the frequency table. It is a figure which shows the data structure of the relevance data. It is a figure which shows the data structure of the analysis conditions. It is a figure which shows the data structure of the station group table 305. 3 is a diagram showing a data configuration of user information 306. FIG. It is a figure which shows the data structure of the station information 307. It is a flowchart which shows the whole processing procedure performed in the area analyzer 1 by 1st Embodiment. It is a flowchart which shows the processing content of step S10. It is a flowchart which shows the processing content of step S20. It is a flowchart which shows the processing content of step S30. It is a figure for demonstrating the base estimation rule used with the area analyzer 1 by 1st Embodiment. It is a flowchart for demonstrating the detail of the process of step S101 which the stay extraction part 101 performs. It is a flowchart for demonstrating the detail of the process of step S102 which the base estimation part 102 performs. 5 is a flowchart for explaining details of processing in step S201 executed by a frequency counting unit 201. It is a flowchart for demonstrating the detail of the process of step S202 which the relevance calculation part 202 performs. 10 is a flowchart for explaining details of processing in step S20201 executed by a single occurrence probability calculation unit 20201. 16 is a flowchart for explaining details of a process in step S20203 executed by the co-occurrence probability calculating unit 20203. It is a figure which shows an example of the station group display screen which the station group display part 303 displays. It is a figure which shows schematic structure of the area analyzer (area analysis system) 1 by the modification of 1st Embodiment. It is a figure which shows schematic structure of the area analyzer (area analysis system) 1 which concerns on the 2nd Embodiment of this invention. It is a figure which shows schematic structure of the area analyzer (area analysis system) 1 by the modification of 2nd Embodiment.

  Hereinafter, an outline of the concept of the present invention will be described first, and then embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, functionally identical elements may be denoted by the same numbers. The attached drawings show specific embodiments and implementation examples based on the principle of the present invention, but these are for understanding the present invention and are not intended to limit the present invention. Not used.

  This embodiment has been described in sufficient detail for those skilled in the art to practice the present invention, but other implementations and configurations are possible without departing from the scope and spirit of the technical idea of the present invention. It is necessary to understand that the configuration and structure can be changed and various elements can be replaced. Therefore, the following description should not be interpreted as being limited to this.

  Furthermore, as will be described later, the embodiment of the present invention may be implemented by software running on a general-purpose computer, or may be implemented by dedicated hardware or a combination of software and hardware.

  In the following description, each information of the present invention will be described in a “table” format. However, the information does not necessarily have to be expressed in a data structure by a table, such as a data structure such as a list, a DB, a queue, or the like. It may be expressed as Therefore, “table”, “list”, “DB”, “queue”, etc. may be simply referred to as “information” to indicate that they do not depend on the data structure.

  In addition, when explaining the contents of each information, the expressions “identification information”, “identifier”, “name”, “name”, “ID” can be used, and these can be replaced with each other. It is.

  When each processing unit (for example, the base data generation unit, the relevance data generation unit, and the area analysis unit) is realized by a “program”, the processor executes the program using the memory and the communication port (communication control device). It will be. For this reason, when describing each processing unit, the processor may be described as an operation subject. Further, the processing realized by the program may be processing performed by a computer such as a management server or an information processing apparatus. Part or all of the program may be realized by dedicated hardware, or may be modularized. Various programs may be installed in each computer by a program distribution server or a storage medium.

(A) Summary of the Present Invention In the present invention, the use history of a traffic IC card, a base station connection log of a mobile phone, or a travel history acquired from a sensor installed in an automobile is used as travel data. The three areas (1) site data generation processing, (2) relevance data generation processing, and (3) area analysis processing are used to analyze the area using the tendency of the movement of the person set. (1) In the base data generation process, a base to be described later is extracted for each user from the movement data. (2) In the relevance level data generation process, the relevance level between locations is calculated for each location based on the number of users based on the location. (3) In the area analysis process, a group in which places having a high degree of association are collected based on the degree of association between places is generated. The outline of each method will be described below. In the following, for the sake of clarity, the usage history of traffic IC cards will be used as movement data, and the station will be described as the location to be analyzed. However, not only the usage history of traffic IC cards but also the movement of mobile phones, cars, etc. You may make it provide a user the result of an area analysis using historical data.

(1) Base data generation process A person moves for some purpose. For example, move to station A, the nearest station for work, move to station B with a large station building for shopping, move to station C, the nearest station for returning home. Then, after staying at the destination for several tens of minutes or several hours, it moves to the next destination. There are not many variations in the purpose of this movement in people's daily lives. In the metropolitan traffic census mentioned above, one of the survey items is “the purpose of travel” of railway users. There are four types of options: “work”, “business”, “private” and “return home”. In addition, there are not so many variations of stations that one person uses in their daily lives, and each seems to be determined to some extent according to the purpose of travel / stay.

  In the present invention, assuming that there are four purposes “home”, “work”, “leisure”, and “outing” as the purpose of staying at the destination in the present invention, a station that tends to be used for each purpose becomes a “base” for that purpose. Think of it as a station. The “base” station varies from individual to individual. If you know your personal home address or work address, you can raise the candidates for “home” and “work” “base” stations, but you can use the first station on a daily basis or return home even if you are a little away. Which station is based, such as preferential use of a station that is convenient for shopping, varies from individual to individual.

  In the present invention, a station that people usually use is extracted from the IC card usage history, and the type of base is estimated from the time zone of travel, the length of time of stay, the frequency of stay, and the like. Specifically, stations that tend to stay for a long time from weekday morning to evening / night are base stations of “work” (including “school”) type, not limited to the day of the week, from evening / night to the next morning Stations that tend to stay are estimated to be "home" type base stations, while stations that tend to stay for a long time on holidays are "leisure" type, and tend to stay for a relatively short time on other holidays and weekdays A certain station is assumed to be a base station of the “outing” type.

(2) Relevance data generation process To analyze the range of people from a certain station, that is, to analyze the power range of a station, what kind of person is coming to the station to be analyzed (destination) It is necessary to know what amount (movement amount) comes from which station (movement source) for the purpose (movement purpose). In addition, it can be considered that stations with a large amount of movement to the analysis target station are included in the power range of the target station.

  Here, if the “movement amount” is “number of people who moved”, the number represents the actual amount of movement from station to station, but depending on the size of the station from which it moved, the dependency between stations is correctly reflected in the power range. There is a problem that is not done. This will be described with reference to FIG.

  FIG. 1 is a diagram for explaining the number of people who moved between stations and the dependency between the stations. In FIG. 1, X is a station to be analyzed, B, C, and D are stations from which a person who has moved to station X moves, and the thickness of the arrow represents the number of people who have moved. As shown in FIG. 1, the number of people moving to X station is large in the order of B station, C station, and D station, but since B station is a large station with many users, the number of people moving to C station and other stations is also large. In many cases, the relationship with station X is relatively weak compared to them.

  On the other hand, although the number of people moving to station X is small compared to stations B and C, station D has the largest number of people as the destination station from station D, and the relationship with station X is the strongest destination. It can be said. In this way, when comparing the strength of the relationship between stations by the number of people who moved, a large-scale station tends to be higher, while a lower-level small station may actually be a more related station. For this reason, for example, when an advertisement for a store in a station building at one station (distribution of coupons, etc.) is deployed at another station, even if it is executed at the source station where the number of people is moving, the cost is higher than expected While the effect cannot be obtained, there is a possibility that a higher effect can be obtained if it is carried out by focusing on a station where the number of moving people is at least strongly related.

  On the other hand, in the present invention, in order to be able to handle the strength of the potential relationship between stations by analysis, the amount of movement between stations is statistically processed to obtain the degree of association between stations. Applying mutual information, AIC, chi-square value, etc. as specific examples of statistical processing, considering the number of people from the source X to the destination other than Y based on the number of people from the source X to the destination Y Find relevance.

(3) Area analysis processing In the present invention, in order to extract a power range of a station such as a trade area or a commuting area for a certain station, a degree of association between stations is used. Specifically, the analysis station as described below is provided to the analyst by extracting a station having a high degree of association with the station as an analysis target station and making it a power range.

(A) Power range extraction means When the analyst specifies the station to be analyzed and the purpose of stay, the power range of the station for the purpose of stay is extracted, and the list of station names included in the power range or on the route map Or display in network format.

(B) Comparison analysis method of power range (i) Comparison by purpose of stay When the analyst specifies a station to be analyzed, the power range is extracted for all the purpose of stay, and a list or route map or Display them side by side in the network format, or display them in different colors on the same route map or network.

(Ii) Comparison between stations When an analyst specifies a plurality of stations and purpose of stay, the power range of the plurality of stations for the purpose of stay is extracted, and a list, route map or network format is provided so that comparison between stations is easy. Are displayed side by side, or are displayed in different colors on the same route map or network.

(Iii) Time series comparison If the analyst specifies multiple stations, purpose of stay, and period for analysis, the power range of the station for the purpose of stay is extracted by period, and a list, route map or Display them side by side in the network format, or display them in different colors on the same route map or network.

(C) Dividing means according to the influence range of the wide area or along the railway When the analyst specifies the wide area, the railway and the purpose of the stay, the station included in the wide area / along is extracted and some of the stations are determined based on the degree of association between the stations. The power ranges are divided into power ranges and displayed in a list, route map, or network format so that the divided power ranges can be easily compared, or displayed in different colors on the same route map or network.

  The outline of the present invention has been described above. Hereinafter, specific embodiments will be described with reference to the drawings.

(B) 1st Embodiment In 1st Embodiment of this invention, a base is extracted for every user using the utilization log | history of a traffic type | system | group IC card, and the degree of association between stations is calculated using the said base. An area analyzer that extracts the power range of a station will be described.

<Overall system configuration>
FIG. 2 is a diagram illustrating a schematic configuration of an area analysis apparatus (also referred to as an area analysis system or an information analysis system) 1 according to the first embodiment. The area analysis apparatus 1 receives the IC card usage history 103 as an input and outputs an analysis report 308 as an output. The IC card usage history 103 is data in which a user of the traffic IC card 81 stores a history of using the traffic IC card 81 with the ticket gate 82 or the checkout machine 83 at the station. The analysis report 308 is a report storing the result of analyzing the power range of the area centered on the station.

  The area analysis apparatus 1 is an apparatus that analyzes an area centered on a station using a movement tendency of a group of people, and is mainly composed of the following three functional units. That is, the base data generation unit 10, the relevance data generation unit 20, and the area analysis unit 30. When the area analysis processing device 1 is configured by a computer, the CPU (processor) included in the computer performs each process of the base data generation unit 10, the relevance data generation unit 20, and the area analysis unit 30 stored in the memory. Are read, and base data generation processing, relevance data generation processing, and area analysis processing are executed, respectively.

(Functional configuration of system: base data generation unit 10)
The base data generation unit 10 extracts a station that is a base of user activity from the movement data of the person. The input is data stored in the IC card usage history 103, and the data is output to the base data 106. The base data generation unit 10 also generates a stay table 104 and a station frequency table 105 as temporary data. Details of these input / output data and temporary data will be described in detail with reference to the drawings in the description of the data structure.

  The site data generation unit 10 further includes two functional units, a stay extraction unit 101 and a site estimation unit 102. Details of these functional units will be described in detail with reference to flowcharts in the description of the processing procedure.

(Functional configuration of system: relevance data generation unit 20)
The degree of association data generation unit 20 calculates the degree of association between stations. The input is data stored in the base data 106, and the relevance data generation unit 20 outputs the data to the relevance data 204 and generates the frequency table 203 as temporary data. Details of these input / output data and temporary data will be described in detail with reference to the drawings in the description of the data structure.

  The relevance level data generation unit 20 further includes two functional units, a frequency count unit 201 and a relevance level calculation unit 202. Details of these functional units will be described in detail with reference to flowcharts in the description of the processing procedure.

(Functional configuration of system: area analysis unit 30)
In the area analysis condition setting unit (area analysis condition setting process) 301, the area analysis unit 30 extracts the analysis condition 304 from the area analysis condition setting information, the user information 306, and the station information 307 input by the analyst. A group of stations is generated according to the conditions and displayed as a station power range. The input of the area analysis unit 30 is data stored in the relevance data 204, and the analysis result is output to the analysis report 308.

  The area analysis unit 30 also generates an analysis condition 304 and a station group table 305 as temporary data.

  Furthermore, the area analysis unit 30 may use data stored in the user information 306 and the station information 307 as reference data. Details of the input / output data, temporary data, and reference data will be described in detail with reference to the drawings in the description of the data configuration.

  The area analysis unit 30 has three functional units, an area analysis condition setting unit 301, a station group generation unit 302, and a station group display unit 303. Details of these functional units will be described in detail with reference to flowcharts in the description of the processing procedure.

  As described above, each functional unit constituting the area analysis unit 30 can be configured by using hardware such as a circuit device that realizes these functions, or an arithmetic device such as a CPU (Central Processing Unit). It can also be configured by a program that defines the operation. In the following, it is assumed that the above functional units are implemented as programs. Data such as the above data, tables, and lists can be stored in a storage device such as a hard disk.

<Hardware configuration>
FIG. 3 is a diagram illustrating a hardware configuration of the area analysis apparatus 1. As shown in FIG. 3, the area analysis apparatus 1 includes a CPU 2, a hard disk (storage device) 3, a memory 4, a display control unit 5, a display 51, a keyboard control unit 6, and a keyboard (input device) 61. And a mouse control unit 7 and a mouse (input device) 71.

  The CPU 2 executes data input / output, reading, storage, and a program in which each functional unit described with reference to FIG. 2 is mounted. The hard disk 3 is a storage device that stores each data described in FIG. The memory 4 is a device that temporarily loads and stores programs and data. The display 51 is a device that displays data to the user, and is controlled by the display control unit 5. The keyboard 61 and the mouse 71 are devices that receive input from the user, and are controlled by the keyboard control unit 6 and the mouse control unit 7, respectively.

<Data structure>
Next, the configuration of each data described in FIG. 2 will be described with reference to FIGS.

(1) Data structure: IC card usage history 103
FIG. 4 is a diagram illustrating a data configuration of the IC card usage history 103. The IC card usage history 103 is data in which a user stores a history of using a traffic IC card, and stores a record of a user touching a card at a terminal installed in a ticket gate or a checkout machine at a station. Has been. In FIG. 4, only the data of one user (U001) is shown. This is because the data is sorted by the user ID, and the history data of a plurality of users are originally stored.

  The IC card usage history 103 has a user ID 10301, a time 10302, a station ID 10303, a terminal type 10304, and an amount 10305 as constituent items. The user ID 10301 is a column for storing the ID of the user of the transportation IC card 81, and the reader / writer device of the ticket checker 82 or the checkout machine 83 reads the user ID stored in the IC card ticket 81. Is obtained by The time 10302 is a column for storing the time when the user uses the ticket gate 82 or the checkout machine 83. The station ID 10303 is a column for storing the ID of the station where the transportation IC card is used. The terminal type 10304 is a column for storing the type of the terminal for which the traffic IC card is used. In the first embodiment, there are three types: “entrance ticket gate”, “entrance ticket gate”, and “checkout machine”. There shall be. The amount 10305 is a column for storing the amount paid by the ticket gate 82 or the settlement machine 83.

(2) Data structure: stay table 104
FIG. 5 is a diagram illustrating a data configuration of the stay table 104. The stay table 104 is data that stores the station where the user stayed, and is generated by the stay extraction unit 101. As shown in FIG. 5, the stay table 104 includes a user ID 10401, a station ID 10402, a stay type 10403, a stay start time 10404, and an end time 10405 as configuration items.

  The user ID 10401 is a column for storing the user ID of the traffic IC card 81. Station ID 10402 is a column for storing the ID of the station where the user stayed. The stay type 10404 is a column for storing the stay type estimated from the start / end time and stay time at which the user stayed. The stay types correspond to the types of bases described above (“home”, “work”, “outing”, “leisure”), and are represented by codes of “H”, “W”, “S”, and “L”, respectively. The stay start time 10404 is a column for storing the time when the stay started, and the end time 10405 is a column for storing the time when the stay ended. In the first embodiment, it is considered that the stay starts and ends with the passage of the ticket gate. Specifically, it is assumed that when a certain station is entered, the previous stay ends and the next stay starts. In general, it is considered that a person leaves the home in the morning and returns to the home at night. Therefore, in this embodiment, the first place of stay on that day is “Home (H)”, and the next stay is switched on the passage (entrance) of the first ticket gate. That is, the first stay at “Home (H)” of the day ends at the passage time of the first ticket gate, and if the next stay is “Work (W)”, the passage time of the ticket gate. Is the start time of stay of “work (W)”. When a user arrives at the nearest station of work and passes (participates) the entrance ticket gate, stays at the place for several hours, and passes (enters) the entrance ticket gate at the same station, "Work (W)" The stay at ends and the next stay begins. As described above, when a stay is extracted from the usage history of the traffic IC card, the start / end time of the stay is the passage (entrance) time of the ticket gate, and the place where the stay is made is the station name (participating station name).

  The stay table 104 is temporary data generated every time the site data generation unit 10 executes processing, and is deleted when the processing of the site data generation unit 10 ends.

(3) Data structure: station frequency table 105
FIG. 6 is a diagram illustrating a data configuration of the station frequency table 105. The station frequency table 105 is data that stores the station where the user stayed and the appearance frequency thereof, and is generated by the base estimation unit 102. As illustrated in FIG. 6, the station frequency table 105 includes a station ID 10501, a stay type 10502, and a stay frequency (for example, indicated by the number of stay days) 10503 as configuration items.

  The station ID 10501 is a column for storing the ID of the station where the user stayed. The stay type 10502 is a column for storing the stay type. The stay frequency 10503 is a column for storing the number of days that the user stayed in the stay type corresponding to the station.

  The station frequency table 105 is generated for each user. The station frequency table 105 is temporary data generated every time the base data generation unit 10 executes processing, and is deleted when the processing of the base data generation unit 10 ends.

(4) Data structure: base data 106
FIG. 7 is a diagram illustrating a data configuration of the base data 106. The base data 106 is data storing a station used as a base by the user, and is generated by the base estimation unit 102. As shown in FIG. 7, the base data 106 includes a user ID 10601, a station ID 10602, a base type 10603, and a stay frequency 10604.

  The user ID 10601 is a column for storing the ID of the user of the traffic IC card. Station ID 10602 is a column for storing the ID of the station used by the user. The base type 10603 is a column for storing the type of station used by the user. In the present embodiment, the type is “H” indicating that the station is the nearest station to the home, “W” indicating that the station is the nearest station to work, and a private destination, such as shopping or a meal. There are four types of “S”, which is a place to stay for a long time, and “L”, which is a private place to go, but a place to stay for a long time, such as leisure facilities and sightseeing spots. To do. The stay frequency 10604 is a column for storing the frequency of staying at the station used by the user.

  The base data 106 is generated every time the base data generation unit 10 extracts a base. Specifically, the system administrator periodically generates site data such as every month or every six months, and additionally stores it in the site data 106.

(5) Data structure: frequency table 203
FIG. 8 is a diagram illustrating a data configuration of the frequency table 203. The frequency table 203 is data storing data calculated as pre-processing for calculating the number of users based on a station and the degree of association between stations, and is generated by the frequency counting unit 201.

  As shown in FIG. 8, the frequency table 203 includes a single frequency table 20300 (FIG. 8A) and a co-occurrence frequency table 20310 (FIG. 8B). The single frequency table 20300 is a table that counts and stores the number of users that are used as predetermined base types (H, W, S, L) for each station. The co-occurrence frequency table 20310 is a table that counts and stores the number of users that are used as combinations of predetermined base types (H and W, H and S, H and L, etc.) for combinations of two stations. The frequency table 203 is temporary data in the relevance level data generation unit 20 and is generated every time the processing of the frequency count unit 201 is executed, and is deleted when the processing of the relevance level calculation unit 202 ends.

  As shown in FIG. 8A, the single frequency table 20300 has a station ID 20301, a base type 20302, a single frequency 20303, and a single occurrence probability 20304 as configuration items. Station ID 20301 is a column for storing the ID of a station. The base type 20302 is a column for storing the base type of the station. The single frequency 20303 is a column for storing the number of users having the station as the base type. The single occurrence probability 20304 is a column for storing the probability that the station appears among all the stations having the corresponding base type. In the single frequency table 20300 in FIG. 8A, the frequency count unit 201 sets values for the station ID 20301, the base type 20302, and the single frequency 20303. The single occurrence probability 20304 is set by the relevance calculation unit 202.

  As shown in FIG. 8B, the co-occurrence frequency table 20310 includes a station ID x 20311, a base type x 20312, a station ID y 20313, a base type y 20314, a co-occurrence frequency 20315, and a co-occurrence probability 20316 as configuration items. Have. Station IDx20311 is a column for storing the ID of a station. The base type x20312 is a column for storing the base type. The station ID y 20313 is a column for storing the station ID. The base type y203014 is a column for storing the base type. The co-occurrence frequency 20315 is that the station stored in the station ID x 20311 is the base type stored in the base type x 20312 and the station stored in the station ID y 20313 is the base type stored in the type of the base type y 20314. This is a column for storing numbers.

  In the present embodiment, the site type x20312 is “home (H)” and the site type x20313 is other than “home (H)”, that is, “work (W)”, “outing (S)”, “leisure (L)”. Any combination is counted as a co-occurrence frequency. That is, among the users who set the station corresponding to the ID stored in the station ID x 20311 to “home (H)”, the station corresponding to the ID stored in the station ID y 20314 is other than “home (H)”. (W) "is counted and stored in the co-occurrence frequency 20315. This co-occurrence frequency corresponds to the number of users who have one station as the nearest station to “home” and the other station as the nearest station to “work”.

  The co-occurrence probability 20316 stores the probability that the combination of the station ID 20311 as the base type x 20312 and the station ID 20313 as the base type y 20314 will occur in all the combinations stored in the co-occurrence frequency table 20310.

  In the co-occurrence frequency table 20310 shown in FIG. 8B, the station ID x 20311, the base type x 20312, the station ID y 20313, the base type y 20314, and the co-occurrence frequency 20315 are set by the frequency counting unit 201. A value of the co-occurrence occurrence probability 20316 is set by the relevance calculation unit 202.

(6) Data structure: relevance data 204
FIG. 9 is a diagram illustrating a data configuration of the relevance data 204. The association degree data 204 is data that stores the association degree between stations, and is generated by the association degree calculation unit 202.

  As shown in FIG. 9, the relevance data 204 includes a station ID x 20401, a base type x 20402, a single frequency x 20403, a station ID y 20404, a base type y 20405, a single frequency y 20406, a co-occurrence frequency 20407, and a relevance degree 20408. And as constituent items.

  Station IDx20401 is a column for storing the ID of a station. The base type x20402 is a column for storing the base type of the station corresponding to the ID stored in the station ID 20401. The single frequency x20403 is a column for storing the number of users who use the station corresponding to the ID stored in the station ID x20401 as the base type stored in the base type x20402. This number is the same as the value of the single frequency 20303 of the single frequency table 20300 of the frequency table 203. Station IDy20404 is a column for storing the ID of a station. The base type y20405 is a column for storing the base type of the station corresponding to the ID stored in the station ID 20404. The single frequency y20406 is a column for storing the number of users who use the station corresponding to the ID stored in the station ID y20404 as the base type stored in the base type y20405. This number is the same as the value of the single frequency 20303 of the single frequency table 20300 of the frequency table 203. The co-occurrence frequency 20407 is that the station corresponding to the ID stored in the station ID x 20401 is the base type stored in the base type x 20402, and the station corresponding to the ID stored in the station ID y 20404 is the base stored in the base type y 20405. This is a field for storing the number of users to be typed. The degree of association 20408 is a station corresponding to the ID stored in the station ID x 20401 calculated based on the value stored in the single frequency x 20403, the value stored in the single frequency y 20406, and the value stored in the co-occurrence frequency 20407. This is a column for storing the degree of association with the station corresponding to the ID stored in the station IDy 20404.

  The relevance data 204 is generated each time the relevance is calculated by the relevance data generation unit 20. Specifically, the system administrator generates the base data periodically such as every month or every six months, and then executes the processing of the relevance data generation unit 20 and additionally stores it in the relevance data 204.

(7) Data structure: analysis condition 304
FIG. 10 is a diagram illustrating a data configuration of the analysis condition 304. The analysis condition 304 is a table that stores area analysis conditions set by the analyst, such as the ID of the station from which the power range is to be extracted and the number of stations belonging to the power range, and is generated by the area analysis condition setting unit 301.

  As shown in FIG. 10, the analysis condition 304 has a central station ID 30401, a base type 30402, a maximum number of related stations 30403, a minimum degree of association 30404, and an exclusive setting 30405 as configuration items.

  The central station ID 30401 is a column for storing the ID of the station from which the power range is extracted. The base type 30402 is a column for storing the base type of the corresponding station. The maximum number of related stations 30403 is a column for storing the maximum number of stations when a station having a high degree of association with the central station is extracted. The minimum relevance 30404 is a column for storing the minimum value of relevance when a station having a high relevance with respect to the central station is extracted. The exclusive setting 30405 is a flag indicating whether the same station may appear in each related station when a plurality of central station IDs are designated (whether each central station may share the same station as a related station). Is a column for storing. Here, “OFF” indicates that exclusive setting is not performed, and that the same station may be a related station for each central station. On the other hand, “ON” indicates that the exclusive setting is performed, and it indicates that only the central station having the highest degree of association among the central stations is a related station.

  In the present embodiment, the analysis condition 304 stores a value designated by the analyst and processed by the area analysis condition setting process 301. The central station ID 30401 is an essential item, and other items are optional. If no analyst is specified, the default value is set. Specifically, the default value of the base type of the base type 30402 is “S”, that is, the “outing” base type, the maximum related station number 30403 is “10”, the minimum relevance 30404 is “−”, that is, “not specified”, The exclusive setting 30405 stores “OFF”, that is, a flag indicating that the same station may appear as a related station for each central station.

(8) Data structure: station group table 305
FIG. 11 is a diagram illustrating a data configuration of the station group table 305. The station group table 305 is a table that stores a list of stations to be analyzed and stations highly related to the station, and is generated by the station group generation unit 302.

  As illustrated in FIG. 11, the station group table 305 includes a central station ID 30501, a base type 30502, and a relevance station ID list 30503 as configuration items.

  The central station ID 30501 is a column for storing the ID of the station to be analyzed. The base type 30502 is a column for storing the type of relationship between the central station and related stations. The related station ID list 30502 is a column for storing a list of IDs of stations having a high degree of association with the analysis target station together with the degree of association.

(9) Data structure: user information 306
FIG. 12 is a diagram illustrating a data configuration of the user information 306. The user information 306 is data that stores user attribute information such as the user's name, sex, and date of birth. In the present embodiment, the usage history of the traffic IC card is used as user movement data. Therefore, the user information 306 stores information on the user of the traffic IC card.

  The user information 306 includes a user ID 30601, a name 30602, a date of birth 30603, a gender 30604, and an address 30605 as configuration items.

  The user ID 30601 is a column for storing the ID of the user of the transportation system IC card. The name 30602 is a column for storing the name of the user. The date of birth 30603 stores the date of birth of the user, and the gender 30604 is a column for storing the gender of the user. The address 30605 is a column for storing the user's address.

(10) Data structure: station information 307
FIG. 13 is a diagram illustrating a data configuration of the station information 307. The station information 307 is data that stores attribute information of stations that appear in the usage history of a traffic IC card.

  The station information 307 includes a station ID 30701, a station name 30702, a route 30703, and an address 30704 as configuration items.

  A station ID 30701 is a column for storing a station ID. The station name 30702 is a column for storing the name of the station. The route 20703 is a column for storing the route name to which the station belongs. An address 30704 is a column for storing the address of the station.

<Processing content of area analyzer>
Next, processing contents of the area analysis apparatus 1 will be described with reference to FIGS.

(1) Contents of Overall Processing FIG. 14 is a flowchart for explaining the entire contents of processing executed in the area analysis apparatus 1 according to the first embodiment.

  The base data generation unit 10 first extracts a station serving as a user's base using the IC card usage history 103 in which the user's movement data is stored (S10).

  Next, the degree-of-association data generation unit 20 calculates the degree of association between stations using the user base data generated in step S10 (S20).

  Finally, the area analysis unit 30 extracts stations that are strongly related to the station specified by the analyst using the relevance data between stations generated in step S20, generates a station group, and displays it to the analyst (S30). ).

  In addition, you may make it perform the process of the base data production | generation part 10 and the relevance degree data production | generation part 20 in this embodiment by batch processing. In the initial state, the processing of the base data generation unit 10 and the relevance data generation unit 20 is executed on a monthly basis for all the IC card usage histories 103 accumulated so far, and thereafter the monthly data is accumulated in that month. The base data generation unit 10 executes the processing for the usage history, stores the base data 106 in the base data 106, calculates the degree of association between stations using the base data of the month, and adds it to the relation degree data 204 Shall be stored. Details of each step will be described below with reference to the drawings.

(2) Process Overview of Base Data Generation Unit 10 FIG. 15 is a flowchart showing a process overview of step S10.

  The stay extraction unit 101 of the base data generation unit 10 extracts the station where the user stayed for each user from the IC card use history 103 and its time and stores them in the stay table 104 (S101).

  Next, the site estimation unit 102 extracts station candidates that serve as sites for each user using the generated stay table 104 and stores them in the station frequency table 105, and selects the type of the site based on the appearance frequency of the station. Estimated and stored in the site data 106 (S102).

(3) Process Outline of Relevance Data Generation Unit 20 FIG. 16 is a flowchart showing a process outline of step S20.

  The frequency count unit 201 of the relevance data generation unit 20 counts the number of users based on the station for each station type for each station. Further, the frequency counting unit 201 counts the number of users who have a certain station as a base other than “Home (H)” type and another station as a base other than “Home (H)” for each combination of base types (S201). ).

  Next, the degree-of-association calculation unit 202 calculates the degree of association between stations based on the counted number of users and stores it in the degree-of-association data 204 (S202).

(4) Process Overview of Area Analysis Unit 30 FIG. 17 is a flowchart showing a process overview of step S30.

  The area analysis condition setting unit 301 of the area analysis unit 30 first generates an analysis condition 304 in accordance with a condition for generating a station and a station group to be extracted from the power sphere input and set by the analyst (S301). More specifically, when the analyst inputs the station names of one or more central stations, the area analysis condition setting unit 301 refers to the station information 307 shown in FIG. 13 and determines the ID of the station corresponding to the target station name. Acquired and stored in the central station ID 30401 of the analysis condition 304. At this time, if the analyst inputs and sets the base type of the target station name, the maximum number of related stations, the minimum degree of association, and the exclusive setting, the area analysis condition setting unit 301 sets the base type 30402 of the analysis condition 304 and the maximum relation, respectively. Stores the values set in the number of stations 30403, the minimum degree of association 30304, and the exclusive setting 30405. Default values are stored for items that are not input by the analyst.

  Next, the station group generation unit 302 generates a station group centered on the designated station according to the set conditions (S302). More specifically, the station group generation unit 302 searches the association degree data 204, and acquires the number of stations having a high degree of association with the station ID stored in the analysis condition 304 by the number specified by the maximum number of associated stations. At this time, stations whose relevance is lower than the value specified as the minimum relevance are excluded from acquisition. If a plurality of central station IDs are designated and the exclusive setting is “ON”, the related station ID list 30503 of each central station is referred to. If the same station is set, the degree of association with the central station Delete the lower one.

  And the station group display part 303 displays a station group to an analyst (S303).

(5) Processing Procedure: Details of Processing of Base Data Generation Unit 10 Next, a detailed processing procedure of the base data generation unit 10 will be described.

(A) Details of Process of Stay Extraction Unit 101 in Base Data Generation Unit 10 FIG. 18 is a diagram illustrating an example of a stay extraction rule used in the stay extraction process. As described above, in this embodiment, four types of stay items “home”, “work”, “leisure”, and “outing” are extracted. In order to extract these stay items, in the present embodiment, a rule is defined that uses three times: a time zone in which the user is estimated to have stayed in the vicinity of the station, its length, and the day of the week. In other words, the stay that appeared at the beginning and end of the day is “home”, “work” if it is other than the beginning and end of the day and 7 hours or more on weekdays, “leisure” if it is “holiday”, otherwise Is “outing”. The site data generation unit 10 extracts stays from the IC card usage history 103 using the rules shown in FIG. 18 as an example, and stores them in the stay table 104.

  FIG. 19 is a flowchart illustrating a detailed processing procedure of step S101 performed by the stay extraction unit 101. In FIG. 19, i is a variable indicating the index of the history stored in the IC card usage history 103. In the present embodiment, it is assumed that the IC card usage history 103 is sorted with only the date at time 10302 as the first key, the user ID 10301 as the second key, and the time at time 10302 as the third key key. That is, the IC card usage histories are first arranged with the same date, and those with the same user ID are arranged in ascending order of time for the same user. In the present embodiment, as will be described later, it is assumed that the last day of the day is 2:59, so when sorting by the date of time 10302, the history up to 2:59 of the next day is handled as the history of the previous day. Also assume that all stored histories are unprocessed. Therefore, when the initial value of i is set to 0, but stays have already been extracted from the history for the past, and stays are extracted from the added IC card usage history, i indicates the index of the added history. . As other variables, Uid is a variable for setting a user ID, and Pid is a variable for setting a station ID, and each is initialized with null. St and Et are variables for setting the start and end times of stay, respectively, and are initialized with null. Further, in the present embodiment, the start of the day is 3 am and the end is 2:59 am the next day. This is because the use history of the traffic IC card is used as movement data, and is combined with the last train time. Hereinafter, each step of FIG. 19 will be described.

(I) Steps S101001 to S101003
The stay extraction unit 101 sets i to 0 (S101001). The stay extraction unit 101 adds 1 to i (S101002). If the user ID 10301 of the i-th usage history of the IC card usage history 103 is the same as Uid, the stay extraction unit 101 skips to step S101006, otherwise goes to step S101004. Proceed (S101003).

(Ii) Step S101004
The stay extraction unit 101 determines that the process has been completed for the daily use history of the user set in Uid, and sets the last time of the day “26:59 (02: 59) "and set" home "stay. Specifically, in the stay table 104, the stay extraction unit 101 sets Uid to the corresponding user ID 10401, sets the Pid value (the ID of the station that participated at the end of the day) to the station ID 10402, and stays. “Home (H)” is set in the type 10403, the St value is set at the start time 10404, and the Et value is set at the end time 10405 (see, for example, the data in the third row in FIG. 5).

(Iii) Step S101005
The stay extraction unit 101 sets the value of the i-th user ID 10301 of the IC card usage history 103 (the ID of the user to be processed next) as Uid, and sets the variable St indicating the start time of the day The first time “03:00” is set and Sv is initialized.

(Iv) Step S101006
The stay extraction unit 101 determines whether i is larger than the number of histories stored in the IC card usage history 103. If it is larger, the stay extraction unit 101 ends the process, and otherwise moves the process to step S101007.

(V) Step S101007
The stay extraction unit 101 determines whether the i-th terminal type 10304 of the IC card usage history 103 is “entrance ticket gate”. If it is “admission ticket gate”, the stay extraction unit 101 shifts the processing to step S101008, otherwise. If there is, the process proceeds to step S101019.

(Vi) Step S101008
The stay extraction unit 101 determines that “stay” has transitioned when the terminal of the usage history is “admission ticket gate” in step S101007, and the i th of the IC card usage history 103 is set in the variable Et indicating the end time of stay. The time stored at time 10302 is subtracted one minute and stored. In other words, this means that it was in a “stay” state before the transition until one minute ago. At this time, if the stay time (length of stay) is calculated from the stay start time St and end time Et and the time is very short (within 15 minutes or the like), it is determined that the transfer is not a stay, You may make it return to step S101002.

(Vii) Steps S101010 to S101101
When the value of St indicates the first stay of the day (St = “03:00”), the stay extraction unit 101 shifts the process to step S101010, and otherwise shifts the process to step S101012 (S101003).

  In S101010, the stay extraction unit 101 acquires the i-th station name 10303 of the IC card usage history 103, refers to the record of the corresponding station information 307, acquires the station ID 30701 of the entrance station, and sets it to Pid. .

  In S101101, the stay extraction unit 101 sets Uid to the user ID 10401 at the end of the stay table 104, sets the value of Pid (station ID of the station that entered the first day of the day) to the station ID 10402, and stay type 10403. “Home (H)” is set to “”, the value set to St is set at the start time 10404, and the value set to Et is set at the end time 10405. In addition, when entering the ticket gate for the first time in a day, it is thought that he stayed at home until just before that. Therefore, here, the previous stay (i-1th stay) is extracted as a stay at home.

(Viii) Step S101012
The stay extraction unit 101 acquires the date from the time 10302 of the IC card usage history 103, and further acquires the day of the week with reference to calendar information (not shown). If the day of the week is a weekday from Monday to Friday, the stay extraction unit 101 shifts the process to step S101013, and otherwise shifts the process to step S101016.

(Ix) Step S101013
The stay extraction unit 101 calculates the stay time (stay length) from the stay start time St and the end time Et. If stay time is more than predetermined time (for example, 7 hours or more), stay extraction part 101 will shift processing to Step S101014, and will shift processing to Step S101015 otherwise.

(X) Steps S101014 and S101015
If you have entered the ticket gate after the second time of the day and have stayed at the previous place for more than 7 hours on weekdays, you may have been working until just before the entrance. . Therefore, in this case, the stay extraction unit 101 extracts the “work (W)” stay as the previous stay (i−1th stay). The stay extraction unit 101 sets the value of each table in the same manner as in step S101011.

(Xi) Step S101015
On the other hand, if you entered the ticket gate after the second time of the day and stayed in the place immediately before it was less than 7 hours on weekdays or less than 4 hours on holidays, until just before the admission It is thought that other general outings. Therefore, in this case, the stay extraction unit 101 extracts the “outing (S)” stay as the previous stay (i−1th stay). The stay extraction unit 101 sets the value of each table in the same manner as in step S101011.

(Xii) Step S101016
The stay extraction unit 101 determines whether or not the stay time is a predetermined time or longer (for example, 4 hours or longer). If the stay time is longer than the predetermined time, the process moves to step S101017. If not, the process moves to step S101015. Transition.

(Xiii) Step S101017
If it was the second time or more of the day that you entered the ticket gate, and you were staying at the previous place for more than 4 hours on a holiday, you thought you were out of vacation until just before the entrance. It is done. Therefore, in this case, the stay extraction unit 101 extracts the “leisure (L)” stay as the previous stay (i−1th stay). The stay extraction unit 101 sets the value of each table in the same manner as in step S101011.

(Xiv) Step S101018
The stay extraction unit 101 sets the i-th time 10302 of the IC card usage history 103 to the variable St representing the stay start time, and shifts the processing to step S101002.

(Xv) Step S101019
The stay extraction unit 101 determines whether the i-th terminal type 10304 of the IC card usage history 103 is “participating ticket gate” or “settlement machine”. If so, the process proceeds to step S101019, and otherwise If so, the process proceeds to step S101002.

(Xvi) Step S101020
When a user enters a ticket gate or passes through a checkout machine, the entry / checkout station becomes the place of stay. Therefore, the stay extraction unit 101 acquires the i-th station name 10303 of the IC card usage history 103, acquires the corresponding station ID 30701 from the station information 307, sets it to Pid, and shifts the processing to step S101002.

(B) Details of processing of base estimation unit 102 in base data generation unit 10 FIG. 20 is a flowchart for explaining the details of the processing of step S102 performed by the stay extraction unit 101. In FIG. 20, i is a variable indicating the number of users (different number) stored in the stay table 104, and j is a variable indicating the number (different number) of stations where one user stays. Hereinafter, each step of FIG. 20 will be described.

(I) Steps S102001 to S102002
The site estimation unit 102 sorts the user data stored in the stay table 104 for each user (unique sort) to create a user list (S102001), and first sets i to 0 (S102002).

(Ii) Steps S102003 to S102004
The site estimation unit 102 adds 1 to i (S102003). If i is smaller than the number of users stored in the user list (not shown), the base station estimation unit 102 moves the process to step S102005, and otherwise performs the process. The process is terminated (S102004).

(Iii) Steps S102005 to S102007
The site estimation unit 102 counts the station where the i-th user stayed for each stay type, and stores it in the station frequency table 105 (S102055).

  Next, the base estimation unit 102 extracts a station having a stay type of H and the highest frequency of H from the stations stored in the station frequency table 105 as H, and stores it in the base data 106 (S102006).

  Then, the base estimation unit 102 extracts, as W, the station having the stay type W and the highest frequency of W from the stations stored in the station frequency table 105, and stores it in the base data 106 (S102007).

(Iv) Step S102008
Subsequently, the site estimation unit 102 creates a unique list of stations stored in the station frequency table 105.

(V) Steps S102009 to S201011
The site estimation unit 102 sets 0 to j (S102009), and then adds 1 to j (S102010).

  Then, the site estimation unit 102 determines whether j is less than the number of stations stored in the station list (not shown) (different number), and if not, shifts the process to step S102012, and otherwise processes. Is shifted to step S201003 (S102011).

(Vi) Step S102012
The base point estimation unit 102 determines whether the base data 106 corresponding to the jth station is stored as H or W. If the base point data is not stored as H or W, the base station estimation unit 102 shifts the process to step S102013. The process proceeds to step S10210.

(Vii) Step S102013
The base point estimation unit 102 determines whether or not the appearance frequency of L is higher than that of S for the j-th station. If the frequency of L is higher, the base station estimation unit 102 moves the process to step S102014, and otherwise performs the process. The process proceeds to step S102015.

(Viii) Step S102014
The site estimation unit 102 extracts the type of the j-th station as L, stores it in the site data 106, and shifts the processing to step S102010.

(Ix) Step S102015
The site estimation unit 102 extracts the type of the j-th station as S, stores it in the site data 106, and shifts the processing to step S102010.

(6) Details of Process of Relevance Data Generation Unit 20 Next, details of the process of the relevance data generation unit 20 will be described using a flowchart.

(A) Details of Processing of Frequency Counting Unit 201 in Relevance Data Generation Unit 20 FIG. 21 is a flowchart for explaining details of processing in step S201 performed by the frequency counting unit 201. Hereinafter, in FIG. 21, i, j, and k are variables indicating indexes of the base data stored in the base data 106. In the present embodiment, it is assumed that the site data 106 is sorted using the user ID as a key. As another variable, Uid is a variable for setting a user ID, and is initialized with null. By the process of FIG. 21, information other than the probability part of FIGS. 8A and 8B is obtained. Hereinafter, each step of FIG. 21 will be described.

(I) Steps S201001 to S201002
The frequency counting unit 201 sets i to 0, j to 1 and k to 0 (S201001).

  Further, the frequency counting unit 201 sets the user ID 10601 of the j-th site data in the site data 106 to Uid (S201002).

(Ii) Steps S201003 to S201004
The frequency counting unit 201 adds 1 to i (S201003), determines whether i is smaller than the number of base data stored in the base data 106, and moves the process to step S201005 if it is small, otherwise processes. Is terminated (S201004).

(Iii) Step S201005
The frequency counting unit 201 determines whether or not the user ID 10601 of the i-th site data of the site data 106 is the same as Uid. If the user ID 10601 is the same, the process proceeds to step S201006. The process moves to S201209.

(Iv) Step S201006
The frequency counting unit 201 obtains the station ID 10602 and the base type 10603 of the i-th base data of the base data 106, refers to the single frequency table 20300 of the frequency table 203, and the base data that matches the station ID 20301 and the base type 20302. Search if already stored. If already stored, the frequency counting unit 201 adds 1 to the frequency of the single frequency 20303 of the record in which the base data is stored. If not, the frequency count unit 201 adds the station ID 10602 to the station ID 20301 at the end of the single frequency table 20300. The value of the site type 10603 is set in the site type 20302, and 1 is set in the single frequency 20303.

(V) Step S201007
The frequency counting unit 201 determines whether or not the i-th site type 10603 of the site data 106 is “H” representing the nearest station to the home, and if “H”, the process proceeds to step S201008, otherwise. If there is, the process proceeds to step S201003.

(Vi) Step S201008
The frequency counting unit 201 sets i to k, and shifts the processing to step S20103. Therefore, the k-th base is always set to “H”. For this reason, in the processing after S201009, the frequency of the combination of the kth base (the base set to H) and the base set to the other category (W, L, S) is calculated. Become.

(Vii) Step S201209
The frequency counting unit 201 determines whether i is the same as j. If i and j are different, the process proceeds to step S201010. Otherwise, the process proceeds to step S201102.

(Viii) Step S201010
The frequency counting unit 201 determines whether j is the same as k. If j and k are different, the process proceeds to step S201011. Otherwise, the process proceeds to step S201012. In the case of j = k, since the co-occurrence frequency of the combination of “H” and “H” is obtained, this combination is excluded in this step.

(Ix) Step S201011
The frequency counting unit 201 acquires the station ID 10602 and the base type 10603 of the k-th base data of the base data 106, the station ID 10602 and the base type 10603 of the j-th base data, and the co-occurrence frequency table 20310 of the frequency table 203. Refer to The frequency counting unit 201 matches the station ID x20311 and the base type x20312 with the station ID 10602 and the base type 10603 of the k-th base data, respectively, and the station ID y20313 and the base type y20314 have the station ID 10602 and the base of the i-th base data. It is searched whether a combination of base data that respectively matches the type 10603 is already stored. If it is already stored, the frequency counting unit 201 adds 1 to the frequency of the co-occurrence frequency 20315 of the record in which the combination of the base data is stored. The station ID 10602 and the base type 10603 values of the k-th base data are set in the station ID x 20311 and the base type x 20312, and the station ID 10602 and the base type 10603 values of the i-th base data are set in the station ID y 20313 and the base type y 20314. Set 1 to the frequency 20315.

(B) Details of Processing of Relevance Level Calculation Unit 202 in Relevance Level Data Generation Unit 20 FIG. 22 is a flowchart for explaining details of the processing in step S202 performed by the relevance level calculation unit 202. The degree of association between stations is calculated according to Equation 1 using the single frequency of the station stored in the single frequency table 20300 of the frequency table 203 and the co-occurrence frequency of the combination of stations stored in the co-occurrence frequency table 20310. In this embodiment, in order to speed up the processing, the single occurrence probability of the station and the co-occurrence probability of the combination of the stations are calculated in advance, and the relevance is calculated by applying the value to the equation shown in Equation 1.

  Hereinafter, each step of FIG. 22 will be described.

(I) Step S20201
The degree-of-association calculation unit 202 calculates the appearance probability of each station for each station type for each station type using the individual frequency for each station type stored in the single frequency table 20300 of the frequency table 203. Details of this processing will be described with reference to FIG.

(Ii) Step S20202
The degree-of-association calculation unit 202 calculates the co-occurrence probabilities of combinations of stations for all combinations using the co-occurrence frequencies of station combinations stored in the co-occurrence frequency table 20310 of the frequency table 203. Details of this processing will be described with reference to FIG.

(Iii) Step S20203
For the combination of stations stored in the co-occurrence frequency table 20310 of the frequency table 203, the relevance calculation unit 202 calculates the single occurrence probability 20304 of the single frequency table 20300 and the co-occurrence probability 20316 of the co-occurrence frequency table 20310 from the above formula. The relevance degree is calculated by applying to 1 and stored in the relevance degree data 204.

(C) Details of Processing of Single Occurrence Probability Calculation Unit 20201 of Relevance Degree Calculation Unit 202 FIG. 23 is a flowchart for explaining details of processing in step S20201 executed by the single occurrence probability calculation unit 20201 of the relevance degree calculation unit 202. It is the flowchart which calculates the probability that a station will appear in a certain base type T. In FIG. 23, T is a variable for setting the type of base to be calculated. In this embodiment, four base types of H, W, S, and L are assumed. Therefore, the process shown in FIG. 23 is repeated for the number of base types by setting the target base type in T. As another variable, i is an index of the frequency data stored in the single frequency table 20300, and TF is a value obtained by totaling the single frequencies 20303 in which the base type 20302 matches T among the frequency data stored in the single frequency table 20300. Is a variable. Hereinafter, each step of FIG. 23 will be described.

(I) Step S2020101
The single occurrence probability calculation unit 20201 sets i and TF to 0.

(Ii) Steps S2020102 to S2020103
The single occurrence probability calculation unit 20201 adds 1 to i (S2020102), and determines whether i is smaller than the number of frequency data stored in the single frequency table 20300 (S2020103). If i is smaller than the number of frequency data stored in the single frequency table 20300, the single occurrence probability calculation unit 20201 shifts the processing to step S2020104. On the other hand, in other cases, it means that the calculation of the single frequency has been completed for all the records, and the single occurrence probability calculation unit 20201 shifts the processing to step S2020106.

(Iii) Step S2020104
The single occurrence probability calculation unit 20201 determines whether the site type 20302 of the i-th frequency data matches T. If they match, the process proceeds to step S202005, and otherwise, the process proceeds to step S2020102.

(Iv) Step S2020105
The single occurrence probability calculation unit 20201 adds the single frequency 20303 of the i-th frequency data to the TF, and shifts the processing to step S2010102. And a process is performed with respect to the following frequency data.

(V) Step S2020106-2020108
The single occurrence probability calculation unit 20201 initializes i to 0 (S2020106), and adds 1 to i (S2020107).

  The single occurrence probability calculation unit 20201 determines whether i is smaller than the number of frequency data stored in the single frequency table 20300. If it is smaller, the process proceeds to step S2020109. Otherwise, the process is performed. The process is terminated (S2020108).

(Vi) Step S2020109
The single occurrence probability calculation unit 20121 determines whether or not the site type 20302 of the i-th frequency data matches T. If they match, the process proceeds to step S201210. Otherwise, the process proceeds to step S202007. Transition.

(Vii) Step S2020110
The single occurrence probability calculation unit 20201 calculates the single occurrence probability based on the single frequency 20303 and TF of the i-th frequency data, stores the single occurrence probability in the single occurrence probability 20304 of the single frequency table 20300, and shifts the processing to step S202007.

(D) Details of Process of Co-occurrence Probability Calculation Unit 20202 of Relevance Degree Calculation Unit 202 FIG. 24 is a flowchart for explaining details of the process of step S20202 executed by the co-occurrence probability calculation unit 20202 of the relevance degree calculation unit 202. It is. i is an index of the co-occurrence frequency data stored in the co-occurrence frequency table 20310, and CF is a variable for setting a total value of the co-occurrence frequencies 20315 stored in the co-occurrence frequency table 20310. Hereinafter, each step of FIG. 24 will be described.

(I) Steps S2020201 to S2020203
The co-occurrence probability calculation unit 20202 sets 0 to i and CF (S202001), and adds 1 to i (S2020202).

  Then, the co-occurrence probability calculation unit 20202 determines whether i is smaller than the number of frequency data stored in the co-occurrence frequency table 20310. If the i is smaller, the process proceeds to step S20204. Shifts the process to step S2020205 (S2020203).

(Ii) Step S2020204
The co-occurrence probability calculation unit 20202 adds the co-occurrence frequency 20215 of the i-th frequency data to the CF, and shifts the processing to step S2020202.

(Iii) Steps S2020205 to S2022307
The co-occurrence probability calculation unit 20202 initializes i to 0 (S2020205), and adds 1 to i (S2020206).

  Then, the co-occurrence probability calculation unit 20202 determines whether i is smaller than the number of frequency data stored in the co-occurrence frequency table 20310. If the i is smaller, the process proceeds to step S2020208, and otherwise. Terminates the processing (S2020207).

(Iv) Step S2020208
The co-occurrence probability calculation unit 20203 calculates the co-occurrence probability based on the co-occurrence frequency 20315 and the CF of the i-th frequency data, stores the co-occurrence probability in the co-occurrence frequency 20316 of the co-occurrence frequency table 20310, and the process proceeds to step S2020206. Let

(7) Screen Example: Example of Station Group Display Screen FIG. 25 is a diagram illustrating an example of a station group display screen displayed by the station group display unit 303 of the area analysis unit 30. In the present embodiment, it is assumed that the analyst can select one of the three display formats shown in FIG. That is, the table format shown in FIG. 25A, the route map format shown in FIG. 25B, and the network format shown in FIG. 25C. Details will be described below.

(I) Table Format Display FIG. 25A shows a screen 303100 that displays a station group in a table format as an example of a station group display screen displayed by the station group display unit 303. As shown in FIG. 25A, the tabular display screen 303100 has a central station 303101, a related type 303102, and a related station 303103 as display areas.

  The central station 303101 is an area for displaying the central station designated by the analyst. Specifically, the central station 303101 is information that is generated and displayed by obtaining a station ID from the central station ID 30501 of the station group table 305 and converting it to a station name with reference to the station information 307. The association type 303102 is information generated and displayed by acquiring a site type from the site type 30502 of the station group table 305 and converting the type code into a site name using the table shown in FIG. The related station 303103 is information that is generated and displayed by acquiring the ID of a related station from the related station ID list 30503 of the station group table 305 and converting it to a station name with reference to the station information 307.

  As shown in FIG. 25A, the tabular display screen 303100 can easily grasp the result in a case where the analyst is familiar with the station name or a case where there are many related stations.

(Ii) Route Map Format FIG. 25B shows a screen 303200 that displays a station group in a route map format as an example of a station group display screen displayed by the station group display unit 303. As shown in FIG. 25B, the route map format screen 303200 has a route 303201, a central station 303202, and a related station 303203 as display items.

  A route 303201 is a route map to which the central station and related stations belong. Specifically, the route 303201 acquires the ID of the central station and the related station from the central station ID 30501 and the related station ID list 30503 of the station group table 305, acquires the name of the route to which the station belongs by referring to the station information 307, The route map is generated and displayed by acquiring the route map. The central station 303202 is displayed by color-coding the central station designated by the analyst. Specifically, the central station 303202 is displayed by acquiring the station ID from the central station ID 30501 of the station group table 305, converting it to a station name with reference to the station information 307, and changing the color for each central station. The related station 303203 acquires the ID of the related station from the related station ID list 30503 of the station group table 305, converts it to a station name with reference to the station information 307, and has the same color as the central station and the degree of relevance with the central station. It is displayed by making the size according to the strength. If the related station has an association with a plurality of central stations, the ratio of the degree of association with each central station is calculated, and displayed in a pie format according to the ratio.

  As shown in FIG. 25B, the route map display screen 303200 can easily grasp the geographical relationship between the stations, so that the power range of the central station can be visually understood.

(Iii) Network Format FIG. 25C shows a screen 303300 that displays a station group in a network format as an example of a station group display screen displayed by the station group display unit 303. As shown in FIG. 25C, the network format screen 303300 has a central station 303301 and a related station 303302 as display items.

  The central station 303301 is displayed by color-coding the central station according to the color designated by the analyst. Specifically, the central station 303301 obtains a station ID from the central station ID 30501 of the station group table 305, converts the station name into a station name with reference to the station information 307, and changes the color for each central station to be a node. Is displayed. The related station 303302 obtains the ID of the related station from the related station ID list 30503 of the station group table 305, converts it into a station name with reference to the station information 307, and has the same color as the central station and the degree of association with the central station. A node having a size corresponding to the strength is displayed, and a link to the central station is also displayed. If the related station is related to multiple central stations, calculate the ratio of the degree of relevance to each central station, display it in nodes that are color-coded in pie form according to the ratio, and display multiple links to the central station To do.

  As shown in FIG. 25 (c), the network display screen 303300 arranges related stations regardless of the distance between stations, and stations that are related to a plurality of central stations are explicitly displayed by links. This makes it easier to understand the competition between stations.

<Analysis parameters>
Using the area analysis apparatus 1 according to the first embodiment, the analyst can specify the following items as analysis parameters (analysis conditions).

(I) Base data generation period In the above description, the base data is generated from the IC card history for one month, but the period is a long period such as six months or one year, a short period such as one week or one day, or Weekdays only, holidays only, and every Wednesday are possible.

(Ii) Relevance level data generation target person In the above description, the relevance level data is generated from the base data of all users who appear in the IC card history, but the target person may be narrowed down by user attributes. Is possible. Specifically, refinement by referring to user information such as user gender (only women, only men), age (20s, 30s, etc.), address (only living in Kanagawa Prefecture), type of base (a station) Narrowing by referring to the base data and the stay table, such as the number of stays (such as those who use “work”) and the number of stays (such as users who have been “outing” at least 4 times a month), or a combination of these It is possible to narrow down (such as women in their 30s who have a certain station as a “work destination”).

(Iii) Frequency used for generating relevance data In the above description, the frequency of single stations and the frequency of co-occurrence between stations are counted by the number of users, but this is not a limitation. Can be weighted by the number of times the station stayed at the station. Specifically, the number of days a user stays at a station may be the single frequency, and the number of days that a user stayed at one station and another station at the same time may be the co-occurrence frequency.

<Example of analysis>
By using the analyzer 1 according to the first embodiment, the analyst can realize the following analysis by specifying the parameters.

(I) Store development strategy analysis for retailers For example, in a certain retail company, the target is women in their 20s and 30s, and from what range the station where the company operates stores It is assumed that there is a need to extract the power range of whether or not you are coming to the station for the purpose of “outing” and verify whether it matches the coverage of the store that the company has assumed for store development.

  In order to respond to this need, analysts use user information to calculate the degree of relevance for women in their 20s to 30s among IC card users, and focus on the station where the company operates stores. What is necessary is just to set area analysis conditions as a station and to extract the power range of a station. Realize efficient store development by expanding / shrinking the store size according to the station's power range if the company's assumed store coverage and the station's power range are different. Can do.

(Ii) Advertising strategy analysis for restaurant companies For example, a certain restaurant company distributes coupon coupons that can be used any number of times at a station near the station where the store is located. Suppose that there is an analysis need to know the range of strength of a station. However, since the purpose of this coupon is to improve the customer repeat rate, it is assumed that there is a condition that the power range is extracted by limiting the target to those who use the railway daily.

  In order to respond to this need, the analyst only needs to calculate the degree of relevance for people who have a high single frequency at the nearest station to extract the power range of the station where the customer store is located. The customer company can improve the effective customer repeat rate by distributing coupons according to the station's power range limited to users who are likely to travel by rail.

(Iii) Analysis of commuting areas for the real estate industry For example, in developing a condominium for a person working in an office district in a real estate company, an analysis of wanting to know the commuting area of the person working in the central office district Suppose there is a need.

  In order to meet this need, the analyst generates base data and relevance data using a long-term IC card history such as one year, and sets the relevant type as the central station at the station located in the downtown area. What is necessary is just to extract the power range of a station as "work". Thereby, the commuting area of the office district can be extracted.

(Iv) Time-series comparative analysis of the area's power sphere In each of the above analysis, analyze the time-series change of the station's power range by changing the accumulation period of the IC card history used when extracting the station's power range. Can do. For example, before or after the opening of a large commercial facility, there is a change in the power range related to “outing” in the surrounding commercial area, and if so, which power range (related station) has changed. It is possible to analyze whether or not there was.

(V) Comparative analysis based on user attributes in the area sphere of influence In each of the above analysis, calculate the degree of association between stations by dividing the target person of the base data used when extracting the power range of the station according to user attributes. Thus, the station power range can be compared and analyzed by user attributes. As an example, by comparing the range of power related to “work” at stations in an office district according to the gender of users, for example, while women's commute area is relatively close to the city center, men commute to the suburbs. Analysis such as spreading is possible.

(Vi) Overhead analysis of a wide area In the above description, the analyst designates the central station and extracts the power range. However, the analysis is not limited to this, and the analyst designates the route or area. It is also possible to divide the station into several groups for all stations included in the route or area.

  Specifically, when an analyst designates a route or a region in setting analysis conditions, the area analysis unit refers to the station information and acquires a station that matches the designated route or address and creates a station list. Next, referring to the frequency table, select the station with the highest probability of single appearance from the stations included in the station list and select it as the central station. Divide into groups. At this time, stations that do not have any relevance to any central station are divided into special groups that do not belong to any central station. In addition, in the route / region designation, it is possible to target all routes / regions. Thereby, a certain route or area can be divided into several groups, and the route or area can be analyzed from a bird's-eye view.

<Modification>
The area analysis apparatus 1 according to the first embodiment generates base data using IC card usage history to generate relevance data, and an analyst analyzes an area according to analysis needs and outputs an analysis report. At this time, there may be a plurality of entities that accumulate and provide IC card usage history, entities that perform analysis, and entities that use analysis reports. FIG. 26 is a diagram showing a system configuration according to a modification of the first embodiment of the present invention.

  As shown in FIG. 26, a plurality of railway companies accumulate the IC card usage history 103, and the analysis service company uses the accumulated IC card usage history 103 to use a standard area using comparatively high analysis conditions. An analysis report is generated by performing an analysis, and a plurality of customer companies can purchase the analysis report regularly. In addition, the entity that provides the service for creating the relevance data may differ from the entity that creates the analysis report by area analysis and provides it to each customer.

(C) Second Embodiment In the second embodiment of the present invention, an analyst analyzes the characteristics of an area, selects a user or a station that is expected to have an effect on content such as an advertisement, and selects the content. A configuration example having a content distribution function to be distributed will be described. Note that the hardware configuration of the area analysis apparatus 1 is the same as that of the first embodiment, and a description thereof will be omitted here.

<Overall system configuration>
FIG. 27 is a diagram illustrating an overall configuration of an area analysis apparatus (area analysis system) 1 according to the second embodiment. The area analyzer 1 has the following four functions. That is, the site data generation unit 10, the relevance data generation unit 20, the area analysis unit 30, and the content distribution unit 91. Since the base data generation unit 10, the relevance data generation unit 20, and the area analysis unit 30 are the same as those in the first embodiment, detailed description thereof is omitted.

  The content distribution unit 91 distributes the content selected by the analyst to the user or station ID narrowed down by the area analysis 30. The distribution condition 92 is data storing conditions for distributing content. For example, when the type of association is “Odekake”, select some of the stations included in the power range of a certain station, make that station the nearest station to your home, and distribute advertising content to users in their 20s and 30s Stores conditions such as

  The content table 93 is data storing content to be distributed. The content 94 is data that is read from the content table 93 by the content distribution unit 91, transmitted to the user's mobile phone 95 or the digital signage 96 at the station, and displayed by these devices. For example, a coupon for a coffee shop can be considered.

  The mobile phone 95 is a mobile phone of the user of the traffic IC card, and its e-mail address is stored in the user information 306.

  The digital signage 96 is an information providing device installed in a concourse of a station, and the installation location is associated with the location stored in the station information 307. That is, it is assumed that an e-mail address is stored in the station information 307, and when the content 94 is transmitted to the address, the content is displayed on the digital signage installed in the station.

  As described above, the area property analysis apparatus 1 according to the second embodiment can distribute content suitable for a user or a place based on the analysis result of the area.

<Modification>
The area analysis apparatus (area analysis system) 1 according to the second embodiment generates base data using IC card usage history to generate relevance data, and an analyst analyzes an area according to analysis needs. Deliver content such as advertisements to users or digital signage at stations, etc., based on the set content distribution conditions and area analysis results. At this time, the entity that accumulates and provides the IC card usage history, the entity that performs analysis, the entity that distributes content, and the entity that requests analysis and distribution of content may be different.

  FIG. 28 is a diagram illustrating a system configuration according to a modification of the second embodiment. As shown in FIG. 28, the railway company stores the IC card usage history 103, and the analysis service company performs area analysis using the IC card usage history 103 stored according to the request of the customer company, and sets the content distribution conditions. Further, the content distribution company can be configured to distribute the content to the user or the digital signage at the station according to the request of the customer company.

(D) Summary (1) The invention made by the present inventor has been specifically described on the basis of the embodiments. However, the present invention is not limited to the above-described embodiments, and does not depart from the spirit of the invention. It goes without saying that various changes can be made. For example, the configuration in one embodiment can be combined or replaced in another embodiment.

(2) In the present invention, the area analyzer includes history data including a movement history of a human group (in the embodiment, it is an IC card use history, but is not limited to this, it is a movement history in a mobile phone or a car) And category data (home, work, outing, leisure, etc.) of a staying place prepared in advance: history data can be classified according to the extraction rules shown in FIG. The staying place of the person belonging to the human group is extracted. Further, the area analysis apparatus obtains the stay frequency for each individual included in the human group in the extracted stay place, and becomes the base of each individual according to the stay purpose prepared in advance based on the stay frequency for each individual. Based on the information on the bases for each staying purpose for each individual, the amount of movement between the places for each staying purpose is tabulated. Then, the area analysis device calculates the degree of association indicating how closely the places by stay purpose are related to each other based on the information on the amount of movement between the places by stay purpose for each individual. Provide relevance information. By doing this, the relationship between each individual's base (the station closest to home or work or the station of the shopping location) (for example, frequently shopping with the station where a user's home is located) Information that is highly relevant to the station where the place is located), and the recipients of this information can take various market strategies.

  Further, not only the information on the degree of association but also information on the power range of the area obtained by processing the information may be provided (area analysis process). More specifically, the area analysis device generates and provides power range information indicating from what range a person is visiting the place to be analyzed based on the information on the degree of association. In this way, the information recipient can immediately use the provided information for the market strategy without further processing from the relevance information.

  Specifically, in the area analysis process, a place, purpose of stay, and period designation are received, and a place with a high degree of relevance with respect to the designated place for the designated stay purpose for a designated period is extracted to influence Range. Alternatively, it receives designations of a plurality of places, extracts places that are highly relevant to the designated places, and sets them as a power range, and displays the power ranges of the specified places on a map or network By displaying in a format, information on the power range may be provided. Alternatively, it receives designations of a plurality of periods, extracts places where the degree of relevance is high in each of the designated plurality of periods as a power range, and displays the power ranges in the specified plurality of periods on a map or in a network format By displaying, information on the power range may be provided. Alternatively, it receives designations of a plurality of stay purposes, extracts places that have a high degree of relevance with respect to the designated stay purposes, and uses them as power ranges, and displays the power ranges for the designated stay purposes on a map or It is also possible to provide information on the power range by displaying it in a network format. By doing in this way, the relationship between each place can be understood more clearly.

  Furthermore, the area analysis apparatus receives the designation of the area, extracts a specific place included in the designated area, and refers to the stay frequency for each individual among the extracted specific places, thereby determining the stay frequency. By setting the high place as the central place and assigning the remaining places to the place with the highest degree of relevance among the central places, the designated area is divided and displayed by the power range. By doing in this way, the information which analyzed the wide area from the bird's-eye view can be provided.

(3) The present invention can also be realized by software program codes that implement the functions of the embodiments. In this case, a storage medium in which the program code is recorded is provided to the system or apparatus, and the computer (or CPU or MPU) of the system or apparatus reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing the program code constitute the present invention. As a storage medium for supplying such program code, for example, a flexible disk, CD-ROM, DVD-ROM, hard disk, optical disk, magneto-optical disk, CD-R, magnetic tape, nonvolatile memory card, ROM Etc. are used.

  Also, based on the instruction of the program code, an OS (operating system) running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing. May be. Further, after the program code read from the storage medium is written in the memory on the computer, the computer CPU or the like performs part or all of the actual processing based on the instruction of the program code. Thus, the functions of the above-described embodiments may be realized.

  Further, by distributing the program code of the software that realizes the functions of the embodiment via a network, it is stored in a storage means such as a hard disk or memory of a system or apparatus, or a storage medium such as a CD-RW or CD-R And the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage means or the storage medium when used.

  Finally, it should be understood that the processes and techniques described herein are not inherently related to any particular apparatus, and can be implemented by any suitable combination of components. In addition, various types of devices for general purpose can be used in accordance with the teachings described herein. It may prove useful to build a dedicated device to perform the method steps described herein. Various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiments. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined. Although the present invention has been described with reference to specific examples, these are in all respects illustrative rather than restrictive. Those skilled in the art will appreciate that there are numerous combinations of hardware, software, and firmware that are suitable for implementing the present invention. For example, the described software can be implemented in a wide range of programs or script languages such as assembler, C / C ++, perl, shell, PHP, Java (registered trademark).

  Furthermore, in the above-described embodiment, control lines and information lines are those that are considered necessary for explanation, and not all control lines and information lines on the product are necessarily shown. All the components may be connected to each other.

  In addition, other implementations of the invention will be apparent to those skilled in the art from consideration of the specification and embodiments of the invention disclosed herein. Various aspects and / or components of the described embodiments can be used singly or in any combination in a computerized storage system capable of managing data. The specification and specific examples are merely exemplary, and the scope and spirit of the invention are indicated in the following claims.

1 Area analyzer (area analysis system)
DESCRIPTION OF SYMBOLS 10 Base data generation part 20 Relevance degree data generation part 30 Area analysis part 91 Content distribution part 92 Content table 101 Stay extraction part 102 Base estimation part 103 IC card utilization history 104 Stay table 105 Station frequency table 106 Base data 201 Frequency count part 202 Relevance calculation unit 203 Frequency table 204 Relevance data 301 Area analysis condition setting unit 302 Station group generation unit 303 Station group display unit 304 Analysis condition 305 Station group table 306 User information 307 Station information

Claims (15)

A storage device for storing history data including a movement history of a human group;
A processor that reads the history data from the storage device and executes predetermined processing;
The processor is
A stay extraction process for extracting stay places of persons belonging to the human group by collating the history data with previously prepared stay place category data;
In the stay location extracted by the stay extraction process, the frequency of stay for each individual included in the human group is obtained, and based on the stay frequency for each individual, the base of each individual is prepared according to the purpose of stay prepared in advance. A site estimation process for estimating the location
Based on the information of the bases for each individual stay purpose, including the amount of movement between the source location and the destination location for each individual for each stay purpose, the source location and the base A frequency counting process that counts the total amount of movement between all the locations
Information on the relative movement amount between the movement source location and the movement destination location for each individual purpose of stay in the total movement amount between the movement source location and all the locations. A degree-of-association calculation process for calculating a degree of association indicating how closely the places by purpose of stay are related to each other;
An information providing process for providing information on the calculated relevance;
An information analysis system characterized by executing. In claim 1,
The processor further executes area analysis processing for generating power range information indicating a range from which a person is visiting a place to be analyzed based on the information on the degree of association, and the power range An information analysis system characterized by providing information. In claim 2,
In the stay extraction process, the processor estimates from the history data by estimating the purpose of stay at the destination based on the location, time zone, and duration of the travel history described in the history data. An information analysis system that extracts the stay location corresponding to the stay purpose. In claim 2,
In the area analysis process, the processor receives a designation of a place, a stay purpose, and a period, and selects a place that is highly relevant to the designated place for the designated stay purpose for the designated period. An information analysis system characterized by extracting it into the power range. In claim 2,
In the area analysis process, the processor receives designations of a plurality of places, extracts places having a high degree of association with the designated places, and sets them as a power range, and the plurality of designated places An information analysis system characterized in that the power range information is provided by displaying the power range on the map or in a network format. In claim 2,
In the area analysis process, the processor receives designations of a plurality of periods, extracts locations having high relevance in the designated plurality of periods, and sets the power range as a power range. An information analysis system characterized by providing information on the power range by displaying the range on a map or in a network format. In claim 2,
In the area analysis process, the processor receives designations of a plurality of staying purposes, extracts places having high relevance with respect to the designated staying purposes, respectively, and sets them as a power range, and the plurality of designated stays An information analysis system characterized by providing information on the power range by displaying the power range related to the purpose on a map or in a network format. In claim 2,
In the area analysis process, the processor receives a designation of an area, extracts a specific place included in the designated area, and refers to a stay frequency for each individual, among the extracted specific places, By setting the place where the frequency of stay is high as the central place and assigning the remaining places to the places with the highest relevance among the central places, the specified area is divided and displayed by the power range An information analysis system characterized by An information analysis method for analyzing history data including a movement history of a human group by a processor and providing the analysis result,
A stay extraction processing step in which the processor extracts the staying places of persons belonging to the human group by comparing the history data with the category data of the staying places prepared in advance.
The processor obtains a stay frequency for each individual included in the person group at the stay location extracted in the stay extraction processing step, and each individual for each stay purpose prepared in advance based on the stay frequency for each individual. A site estimation process step for estimating the location of the site,
The source location including the amount of movement between the source location and the destination location for each individual by the purpose of stay based on the information of the base for each individual purpose of stay by the processor And a frequency count processing step for totalizing the total amount of movement between all the locations serving as the bases,
Relative movement between the source location and the destination location for each individual purpose of stay in the total amount of movement between the source location and all locations. A relevance calculation processing step for calculating a relevance indicating how closely the places according to the purpose of stay are related to each other based on the amount information;
An information providing processing step in which the processor provides information on the calculated relevance;
An information analysis method comprising: In claim 9,
Further, the processor executes area analysis processing for generating power range information indicating how much a person is visiting the place to be analyzed based on the information on the degree of association, and the power range An information analysis method comprising an area analysis processing step for providing the information. In claim 10,
In the stay extracting process step, the history data is estimated by estimating a purpose of stay at a destination based on the location, time zone, and duration of the travel history described in the history data. The information analysis method, wherein the stay location corresponding to the stay purpose is extracted from. In claim 10,
In the area analysis processing step, the processor receives a designation of a place, a stay purpose, and a period, and a place having a high degree of association with the designated place in the designated stay purpose for the designated period An information analysis method characterized by extracting the power into the power range. In claim 10,
In the area analysis processing step, the processor receives designation of a plurality of places or a plurality of periods, and extracts a place having a high degree of association with the designated plurality of places or a plurality of periods, respectively, and a power range An information analysis method characterized in that the power range information is provided by displaying the power ranges of a plurality of designated locations or a plurality of periods on a map or in a network format. In claim 10,
In the area analysis processing step, the processor receives designations of a plurality of staying purposes, extracts places having a high degree of association with respect to the designated staying purposes, and sets the power ranges as the power ranges. An information analysis method characterized in that information on the power range is provided by displaying the power range related to the purpose of stay on a map or in a network format. In claim 10,
In the area analysis processing step, the processor receives a designation of an area, extracts a specific place included in the designated area, and refers to a frequency of stay for each individual among the extracted specific places. By setting the place where the stay frequency is high as the central place and assigning the remaining places to the places with the highest relevance among the central places, the specified area is divided by the power range An information analysis method characterized by displaying.

Images