JP5823332B2 - Personal behavior preference analysis system and personal behavior preference analysis method - Google Patents

Description

  The present invention relates to a method and an apparatus for preferentially presenting a route suitable for each user, such as time priority and charge priority, by estimating personal preference for a travel route from a traffic IC card usage history. is there.

  When heading for the first place using public transportation, a method for searching for a route from a departure point to a destination using a route search system or the like is widely used. In train transfer guidance by a general route search system, the order of presentation of search results is determined according to the original criteria set by the company that provides the route search system, so the order of arrival time or the order of cheapness is not necessarily, In many cases, the order of presentation was not clear. Therefore, if the request for the route is clear, such as if you want to get to your destination as soon as possible, or if you want to go on the cheapest route, the user will browse all the presented routes and rearrange them. And had to decide which route to choose.

  By the way, with advances in wireless sensor devices, GPS, non-contact IC technology, etc., the position and behavior data of people and objects are acquired in various industrial fields, and behavior history information extracted by analysis is used for marketing and the like. I came. For example, since an ID code for identifying each card is assigned to a transportation IC card ticket, by associating the location of the ticket gate where the ticket passes and the store where the ticket was purchased, It becomes possible to grasp the behavior of the owner to some extent.

  On the other hand, Patent Document 1 discloses a system that provides an accurate and optimum transportation means to a destination via a portable information terminal owned by the user to a transportation means user. Further, Patent Document 2 presents the user with output route data obtained by searching for transportation changes from the departure point to the destination, and supplementary route data obtained by searching for changes in the same section under other different conditions. A system is disclosed. Further, Patent Document 3 discloses a system that provides information that suits the user's preference by using the history of transportation payment means in providing information such as recommendations for purchase behavior in and around the station. Is disclosed.

Japanese Patent Laid-Open No. 2006-234813 JP 2009-51293 A JP 2011-76291 A

  By the way, the technique described in Patent Document 1 presents a route that is considered to be most appropriate to the judgment of the public to the user, and therefore does not present it for each user in consideration of palatability. In addition, the technique described in Patent Document 2 presents a route after comparing a plurality of routes, which is an effective clue to the user when selecting a route. However, preference is given to each user. I don't remember it. Furthermore, the technique described in Patent Document 3 is a recommendation method for purchasing behavior, not for moving behavior.

  The present invention has been made in view of such a point, and its purpose is mainly for railway organizations from history information (train getting-on / off data) of traffic IC cards used daily by a large number of people. By extracting the most appropriate route judged by the masses and comparing the result with the conditions of the route actually used by each user, the preference of the user's movement behavior is estimated, and the user's behavior It is to provide information reflecting trends.

The above-mentioned problems are: log data with an ID including use type information indicating entry and participation, log ID, user ID, station ID, and use time, route ID, entry station ID, entry station ID, fare information, and required time. A data storage unit for storing a route master including the number of transfers, a station area definition list including an area ID, a representative station ID, and a plurality of station IDs; and the log data with ID stored in the data storage unit; based on the said route master, and the log ID and the user ID and entry date and played date and travel log generating portion that the travel log multiple production comprising the entrance station ID and the participating station ID and the route ID, the Based on the admission date and time and the participation date and time and a predetermined connection determination threshold of a plurality of generated movement logs, a movement log connection unit that connects the plurality of movement logs to each other, and the data Based on the route master stored in the storage unit and said station area definition list, to generate a section route list including said path ID and segment ID and departure area ID arrival area ID and the number of routes, the The required time, the fare information, and the number of transfers for the section ID are calculated based on the number of routes in the section route list and the route master, and the required time rank, the fare information rank, and the number of transfers for the section ID are calculated. calculated as a representative route calculation unit for generating a route ranking list including said segment ID wherein the departure area ID the arrival area ID and the required time order and said fare information order and the transfer count order, the Based on the generated route order list, a required time priority, a fare information priority, and a transfer frequency priority are generated for each user ID, and the user ID and the place An action preference determination unit that generates a selection criterion determination result list including a time priority, the fare information priority, and the transfer frequency priority, and accepts user input, based on the selection criterion determination result list, extract the required time priority and fare information priority and the number of times of transfer priority corresponding to the input of the accepted user, the said duration priorities issued extracted and the fare information priority and the number of times of transfer priority It can be solved by an individual behavior preference analysis system having an interface unit that displays output information based on the screen.

As a means for solving the above problems, in the present invention, using the usage history of a payment terminal such as a traffic non-contact IC card or a mobile phone having an equivalent function, the behavior preference of a station user during movement is improved. It is characterized by estimating.
Furthermore, the present invention calculates the above-mentioned user's behavioral preference according to time zone or season, and based on this, information analysis distribution that preferentially presents a route that matches the user's preference at the time of route search Device.

According to the present invention, since data of all users who possess a traffic IC card is used, generally, when obtaining an appropriate route, more reliable without being biased to data of some users. High results can be obtained.
In addition, according to the present invention, when calculating the user's behavior preference when moving, it can be calculated in detail such as weekdays, holidays, time zones, seasons, etc., so the estimation result is stored and the user can search for a route. When using the system, it is possible to provide a program that returns the route search results that meet the requirements.
In addition, according to the present invention, by summing up users who use the station for each behavioral preference and comparing by station unit, the difference between the stations and routes is finely indexed such as weekdays, holidays, time zones, seasons, etc. It becomes possible to become.

It is the schematic explaining the relationship between an area and a path | route. 1 is a basic configuration diagram of an entire system for implementing the present invention. It is a figure explaining the structure of the record which stores the utilization log | history information of a traffic type | system | group IC card. It is a figure explaining the structure of the record which stores the basic information of a station, a route, and a route. It is a figure explaining the structure of the record which stores the movement log produced | generated from the utilization log | history of a traffic type | system | group IC card, and the data after the connection of a movement log. It is a figure which shows the process sequence which produces | generates a movement log from the utilization log | history of a traffic type | system | group IC card. It is a figure which shows the process sequence which connects the movement log produced | generated from the utilization log | history of a traffic type | system | group IC card. It is a figure which shows the structure of the record which stores the definition of a station and an area. It is a figure which shows the structure of the record which refers a path | route master and stores the path | route about all the areas. It is a figure which shows the process sequence which refers a path | route master and extracts a path | route about all the sections. It is a figure which shows the structure of the record which stores the path | route condition order | rank of each path | route in a multiple path | route area. It is a figure which shows the process sequence which calculates the path | route condition order | rank of each path | route in multiple path | route area. It is a figure which shows the example which graphed the route condition order | rank of each user's movement route. It is a figure which shows the example in which individual preference appears in distribution, when the route condition order | rank of each user's movement route is graphed. It is a figure which shows the structure of the record for storing the determination result of the action preference at the time of a movement for every user. It is a figure which shows the process sequence which determines the action preference at the time of a movement for every user. It is the figure which showed the example of a screen of the transfer guidance map which shows a path | route for every selection preference. It is a figure which shows the result of having totaled the user who uses each station for every selection preference. It is the figure which showed the example of the transfer route search screen for users. It is the figure which showed the example of the search result screen of the transfer route search engine for users.

  FIG. 1 to FIG. 20 show an example of a method for determining an individual's behavior preference when moving according to the present invention and a system for providing information to a user or a company.

  FIG. 1 schematically shows the relationship between stations, routes, sections and routes in a railway network. In general, in the railway network, there are many so-called transfer stations in which stations of a plurality of railway companies are close to each other. For example, station 1 (001) and station 2 (002) are adjacent to station 3 (003), station 4 (004) and station 5 (005), and station 6 (006) and station 7 (007) are adjacent to each other. And exist. From station 1 (001), use route 1 (011) to go to station 3 (003) and route 2 (012) to station 6 (006), then station 7 (007) And there is a route 2 that arrives at the station 4 (004) using the route 3 (013), and a route 3 that arrives at the station 5 using the route 4 (014) from the station 2 (002). Suppose there is. When the user wants to move from the vicinity of the station 1 to the vicinity of the station 3, the options of the station 1 (001) and the station 2 (002) are selected for the departure area A, and the station 3 (003) is selected for the destination area B. ), Station (004), station (005), and there are three options of route 1 to route 3 between them. In this way, a line connecting a certain departure area and destination area is defined as a section, and a route connecting stations between the stations is defined as a route.

  FIG. 2 is an overall configuration diagram of the information processing system and the estimation of the user's behavior preference during movement according to the present embodiment. The main components of the system are a non-contact type IC card (or a portable terminal having an equivalent function) 103 used by the user 101, and a ticket gate for use in a railway station, around the station, or in a store. And a network 104 for transmitting data acquired by those ticket gates to a server group. The server group includes a history server 111, a calculation server 112, and an information distribution server 113, and performs accumulation and analysis processing of usage data of the non-contact type IC card 103. In addition, description is abbreviate | omitted about the function and structure in a non-contact IC card and a ticket gate which are not directly related when describing this invention.

  When the user 101 holding the non-contact type IC card 103 passes the ticket gate 102, the user ID for identifying the IC card 103 and the location information including the passage date and time are stored in the ticket gate 102, and the server is used as the original data. 105 is accumulated. At the same time or at an appropriate timing such as every other hour or every other day, necessary parts are transmitted to the history server 111 via the network 104. The same applies to a card reading terminal, a POS cash register terminal, a vending machine, etc. installed in a store other than the ticket gate, for example, in the station premises or around the station. For example, when the user 101 holding the non-contact type IC card 103 holds the card over the card reading terminal in the store, the purchase information including the user ID for identifying the IC card 103, the date and time, the amount of money, etc. is stored in the card reading terminal. And stored in the server 105 as original data. At the same time or at an appropriate timing such as every other hour or every other day, it is transmitted to the history server 111 via the network 104.

  Each server group of the history server 111, the calculation server 112, and the information distribution server 113 is connected to the network 104, and can communicate with each other or with each ticket gate 102 and user (115, 117). In this embodiment, the server group of the history server 111, the calculation server 112, and the information distribution server 113 will be described. However, it is possible to configure the server group so that the functions of these server groups can be executed by one or a plurality of servers. .

  The history server 111 receives user data read by an IC card reader terminal such as a ticket gate machine from the original data server 105 through the network 104 and records it in the data storage unit 121 in the server. The data to be collected and stored includes log data 122 with ID for storing the position history of each user, basic master data 123 related to stations and routes, and the like. Further, movement log data 124 obtained by first processing the ID-added log data 122 and connected movement log data 125 subjected to processing for connection to the movement log data 124 are stored. The basic master data 123 and the like related to the station and the route are appropriately input from the outside of the system and updated / recorded when there is a change or an update.

  The calculation server 112 uses the data stored in the history server 111 to generate movement data for the IC card user and to calculate the movement time for each section or in the station premises. The calculation server 112 mainly includes a network interface (I / F (A) 139), a CPU 131, a memory 132, and a storage unit 133. The network interface is an interface for connecting to a network. The storage unit 133 stores a program group such as a movement log generation program 134, a movement log connection program 135, a representative route calculation program 136, and an action preference determination program 137, and statistical values and index values obtained as a result of calculation processing. A data storage unit 138 for storing is included. The storage unit is, for example, a hard disk drive, a CD-ROM drive, or a flash memory. Various programs and various data may be divided and recorded on a plurality of recording devices.

  When each program group is executed, data to be analyzed is read from the history server 111 and temporarily stored in the memory 132, and the CPU 131 reads each program (134, 135, 136, 137) into the memory and executes it. By doing so, various functions are realized. The execution timing of these programs may be performed, for example, every time a new request is added to the user's request timing or the history server 111, or automatically as a batch process at a predetermined time every day. May be done.

  The information distribution server 113 includes a network interface (I / F (B) 143 and I / F (C) 144), a CPU 145, a memory 146, and a recording device 147. The network interface is an interface for connecting to a network. The recording device records various programs and various data, such as a hard disk drive, a CD-ROM drive, and a flash memory. Various programs and various data may be divided and recorded on a plurality of recording devices.

  The information distribution server 113 allows a user (115, 117) to search for a route that matches his / her preference from the portable information terminal 116 or the home or public information terminal 118 via the Internet 114, This is for referring to information such as evaluation by user's preference. The contactless IC card 103 owned by the user (115, 117) is held over the IC card reader 119, transmitted to the information distribution server 113 via the Internet 114, and handed to the search processing program 141. May be specified. A search processing program 142 and an information distribution program 143 are recorded in the recording device. The CPU executes various functions by reading various programs recorded in the recording device into a memory and executing them. Specifically, by executing the search processing program, the data search is performed based on the conditions input by the user, and the information is processed to be presented to the user by executing the information distribution program. Distribute information. These pieces of information are basically acquired at the timing when each user actively accesses.

  FIG. 3 is a diagram showing one record of the data structure of typical data stored in the history server 111. First, the log data 122 with ID is a log ID 241, a target user ID 242, a station ID 243 associated with information on which data reading terminal has passed, a use time 244 that has passed through the reading terminal, and whether it is entry or exit. Information such as the usage type 245 of. Here, the usage type is information indicating the type of processing such as “entrance” or “participation” for a ticket gate or an entrance gate, and “purchase” for a product sales terminal, for example. The ID-added log data 122 may be sent each time new data is generated, or may be sent collectively at midnight when usage is reduced. On the history server 111 side, the storage process may be performed in accordance with the transmission timing.

  FIG. 4 is a diagram showing the types of master data 123 stored in the history server 111 and their data structures. First, the station master 260, which is basic data regarding a station, includes information such as a station ID 261 for identifying a station, a station name 262, a railway company 263, a location 264 such as an address, and latitude / longitude information 265. When there is a change in the station or route configuration, data is added or modified as needed. The route master 270, which is basic data related to the route, includes information such as a route ID 271 for identifying a route, a route name 272, a railway company 273, and a route type 274 for distinguishing between a normal train and an express train. The station / route relation master 280, which is basic data for associating a station and a route, has a route ID 281 for identifying the route, a station ID 282 included in the route, a sequence number 283 for managing the order of the station, a station that stops, Information such as a station type 284 for identifying which station is passing and a required time 285 from the first station is included. Further, the route master 290, which is basic data regarding the route, includes route IDs 291 and entrance station IDs 292 for identifying routes, information on route IDs and transfer station IDs for the number of routes to be boarded, in addition to route IDs 293. In addition, information on a period 294 and a time zone 295 during which this route is valid is included. Information about the period and overtime may be set at predetermined intervals, such as every month or every hour, or may be set according to the schedule revision schedule or the train operation mode. Next, when going from the entrance station ID 292 to the exit station ID 293, the data is stored in the route ID 1 (296) for identifying which train to take when taking the train only once. In addition, when going from the entrance station ID 292 to the exit station ID 293, when taking the train more than once, the route ID 1 (296) for identifying which train to ride, the transfer station ID 1 (297) for identifying the transfer station, and the next train to ride Is sequentially stored in accordance with the value of the number of times of boarding 301, such as route ID 2 (298). Also included are information such as the number of times of boarding 301 and the standard time 302, the shortest time 303, and the fare 304 indicating the comprehensive information of this route. Here, there may be a plurality of routes corresponding to the combination of the entry station ID 292 and the entry station ID 293, but here, the route with the highest use frequency is generally assigned as the first route. These master data 123 are input, updated, and recorded from the outside of the system shown in FIG. 2 each time the station or route is changed, for example.

  FIG. 5 is a diagram showing a data structure for storing the movement log data 124 stored in the history server 111 and the post-connection movement log data 125 generated by connecting the movement log data. The movement log data 124 includes a log ID 321 for identifying a log, a target user ID 322, an entry date and time 323 indicating the time of entry to the station through the ticket gate from outside the station, and passing through the ticket gate from the inside of the station. Entry date / time 324 indicating entry time, entry station ID 325, entry station ID 326, payment amount 327 indicating the fee for movement, route ID 328 associated with route information included in the route master 290, entry date / time 323 Information such as a purchase flag 329 indicating whether or not there has been a purchase with the same IC card is included. The movement log data 124 is data after primary processing generated from the log data 122 with ID. The post-connection movement log data 125 includes a log ID 331 for identifying a log, a target user ID 332, an entry date / time 333 indicating the entry time of the first movement log, an entry date / time 334 indicating the entry time of the last movement log, the first Information such as a departure station ID 335 indicating the entry station of the movement log, an arrival station ID 336 indicating the departure station of the last movement log, an ID 337 of the movement log serving as the start point, and an ID 338 of the second movement log are included.

  FIG. 6 is a diagram for explaining a procedure for generating the movement log data 124 from the ID-added log data 122 that is the usage data of the traffic contactless IC card 101 and storing it in the history server 111. Here, a description will be given on the assumption that the storage processing in the history server 111 is performed by batch processing once every predetermined time every day. First, with reference to the user ID 242 and the use time 244 included in the newly collected log data 122 with ID, all data are rearranged in order of user ID and time (processing step 400). Next, the following same process is repeated for the number of user IDs for the data rearranged in process step 400 (process step 401). First, the variables corresponding to the entry station, entry date / time, entry station, entry date / time, and purchase flag are initialized (processing step 402). Next, case-by-case processing is performed on the data arranged in order of time according to the value of the usage type 245 (processing step 403). If the value of the usage type 245 is entrance (processing step 404), the value stored in the station ID 243 is set in the variable of the entrance station, and the value stored in the usage time 244 is set in the variable of the entrance date / time. 0 is assigned to the variable of the purchase flag (processing step 405).

  On the other hand, when the value of the usage type 245 is purchase (processing step 410), if the entry station and the entry date and time variables are set, 1 is substituted into the purchase flag variable and the value is set. If not, the data is ignored and the processing proceeds to the next data (processing step 411).

  Here, when the value of the usage type 245 is participation (processing step 406), the value stored in the station ID 243 as the entry station variable only when the entry station and entry date and time variables are set. And the value stored at the use time 244 are set in the variable of the participation date (processing step 407). Further, the corresponding route is searched from the route master 290 using the values of the two variables of the entrance station and the exit station, and the matched route ID 291 is extracted (processing step 408). Finally, the log ID, user ID, entry date / time, entry date / time, entry station ID, entry station ID, payment amount, and route ID information are stored as one record in the movement log data 125 (processing step 409). Here, the log ID 321 is held as a serial number.

  FIG. 7 illustrates a procedure for generating post-connection movement log data 125 in which the movement log data 125 is connected according to time rules and storing the movement log data 125 in the history server 111 in order to extract the user's real departure point and destination. FIG. Similarly to the generation of the movement log data, the post-connection movement log data generation process will be described assuming that the storage process in the history server 111 is performed once a batch at a predetermined time every day. First, all the data of the movement log data 124 generated from the newly collected log data 122 with ID is rearranged in order of user ID and time with reference to the log ID 321 and user ID 322 (processing step 500). Next, the following same process is repeated for the number of user IDs for the data rearranged in process step 500 (process step 501). First, variables corresponding to the departure station, departure date / time, arrival station, and arrival date / time are initialized, and a threshold Dt relating to a standard transfer time for determining whether or not connection is possible is set (processing step 502). The threshold Dt regarding the standard transfer time is a positive value, and may be a value common to all railway networks, or a different value may be provided for each area. Next, the following same processing is repeated for the data arranged in time order (processing step 503). If any of the departure station, departure date / time, arrival station, arrival date / time is not set (processing step 504), the entry station ID and the entry date / time of the movement log are set in the departure station and departure date / time, and the arrival station and arrival date / time. Substitute the station ID and the date and time of entry in the travel log (processing steps 505 and 506). When the departure station, departure date / time, arrival station, and arrival date / time have already been set (processing step 507), the difference between the arrival date / time and the entry date / time of the current travel log is obtained, and the difference is related to the standard transfer time. If it is smaller than the threshold value Dt, it is considered that the movement was merely a transfer and was not the original destination. Therefore, it is assumed that the connection is possible, and the entry station ID and entry date / time of the current movement log are substituted into the arrival station and arrival date / time. (Process Step 508). In addition, the log ID of the movement log is held. Then, in order to determine whether or not the next movement log can be connected, the process proceeds to the next data processing. On the other hand, the difference between the arrival date and time and the entry date and time of the current travel log is obtained. If the difference is larger than the standard transfer time threshold Dt, it is determined that some purpose has been completed at the station and it is determined that the connection cannot be made. Here, the record that has been disconnected is moved after connection with a list of connected log IDs 331, user IDs 332, departure date and time 333, arrival date and time 334, departure station ID 335, arrival station ID 336, and stored movement log IDs. Stored in the log data 125. Also, the entry station ID, entry date / time, entry station ID, and entry date / time of the current travel log are substituted into the departure station, departure date / time, arrival station, and arrival date / time variables, and the processing proceeds to the next data (processing step 509). . Here, the log ID 331 is held as a serial number.

  FIG. 8 is a diagram showing a data structure for storing station-area definition data for regarding the group of stations that can be transferred described with reference to FIG. 1 as the same area. The station-area definition list 340 includes an area ID 341 for identifying a record, a representative station ID 342 included in the area, a target period 343 in which the definition data is valid, the number of stations 344 included in the area, an area Station ID1 (345), station ID2 (346), station ID3 (347), station ID4 (348), etc. contained in the inside are contained.

  FIG. 9 is a diagram showing a data structure for storing result data obtained by counting how many routes exist in each section. The section-route list 350 includes a section ID 351 that identifies a section, a departure area ID 352 that indicates a departure place, an arrival area ID 353 that indicates an arrival place, a target period 354 that is an aggregation target, a time zone 355 that is an aggregation target, Information such as the number of routes 356, the first route ID 357, and the second route ID 358 included in this section is included. This list includes the IDs of all routes in the section.

  FIG. 10 is a diagram showing a processing procedure for extracting paths that can exist for all sections and storing them in the section-path list 350. First, the route master 290 refers to the period 294 and the time zone 295 to extract all data corresponding to the preset target period and time zone (processing step 600). Here, it is assumed that the value of the target period and time zone is set in advance externally, and the same processing is performed for the combination of all the set data target time periods and time zones. An example of processing when the target period and time zone are determined is shown. The following processing is repeated for all the extracted route data (processing step 601). Refer to the entry station ID 292 and entry station ID 293 from the route master data, and from the station-area definition list 340, the departure area ID and arrival of which area the first entry station and the last entry station of the route belong to The ground area ID is extracted (processing step 602). It is searched whether a record corresponding to the combination of the departure area ID and the arrival area ID already exists in the section-route list 350. If there is a record (processing step 603), the record of the corresponding section ID is included in the record. A new route ID is added, and 1 is added to the value of the number of routes (processing step 604). If a record corresponding to the combination of the extracted departure area ID and arrival area ID does not already exist in the section-route aggregation list 350 (processing step 605), a section ID is assigned and the number of routes is set to 1 Store as a new record in the route list 350 (processing step 606).

  FIG. 11 shows a section-path total list 350 in which the number of usages is totaled for each section and route shown in FIG. 9, and a plurality of routes exist in one section, and conditions such as a fare and a required time are included in each route. It is the figure which showed the data structure for storing the data regarding a several path | route area which has a difference. The route order list 370 includes a section ID 371 for identifying a section, a departure area ID 372 indicating a departure place, an arrival place ID 373 indicating an arrival place, a target period 374 to be counted, a time zone 375 to be counted, and a path 376. , The required time interval ranking 377 in the order of the required time in the route included in this section, and the fare interval ranking 378 in the route included in this section in ascending order of the fare. In addition, the information includes information such as the intra-change interval rank 379 in which ranks are ranked in ascending order of the number of transfers in the route included in this interval. Here, an example of storing the ranking results for the three route conditions has been shown, but other factors such as the number of trains and the variance of the required time can be considered. Further, not only the route master data but also the utilization rate based on the actual condition of the user may be calculated and ranked.

  FIG. 12 is a diagram showing a processing procedure for extracting data from the section-route list 350 and storing it in the route order list 370. First, all records in the section corresponding to the target period and time zone set in advance from the section-path list 350 and having a number of paths 356 greater than 1, that is, a section having a plurality of paths are extracted (processing step 700). . Here, it is assumed that the value of the target period and time zone is set in advance externally, and the same processing is performed for the combination of all the set data target time periods and time zones. An example of processing when the target period and time zone are determined is shown. The following processing is repeated for all the extracted records (processing step 701). For all routes included in the current record (processing step 702), the route master 290 is referred to and conditions such as required time, fare, number of transfers are extracted (processing step 703). Based on the extracted conditions, the following processing is repeated for all routes (processing step 704). The required time, fares, and number of transfers are sorted in ascending order of value to obtain the rank within the section (processing step 705) and stored in the route ranking list 370 (processing step 706).

  FIG. 13 shows an example in which the personal preference can be expressed as a distribution bias when the order of the route conditions actually traveled by each user is displayed in the route of the section. It is a figure. For example, the required time, the fare, and the number of transfers are each axis, and the required time interval rank 377 and the fare interval rank 378 are recorded for each user from the records stored in the route rank list 370 in the three-dimensional space. If the data of the rank 379 in the number of transfers section is referred to and mapped, it is considered that the data is concentrated in the area where the priority of the priority elements is low. For example, the distribution of users who give priority to fares varies greatly in the order of required time and number of transfers. Since the fluctuation range of the maximum value and the minimum value is different for each condition, it is necessary to normalize in consideration of them, but by calculating the variance using the normalized value and comparing the size, You can specify or narrow down which conditions are prioritized.

  FIG. 14 shows the result of mapping data in the multidimensional space as shown in FIG. 13, narrowing down the user's priority conditions to two, and using only the top two condition ranking results, It is the figure which showed the case where a person's preference appears more clearly in the distribution by making a graph on the plane, taking the two axes of fare and required time as an example. It can be said that the route having the highest route order stored in the route order list 370 is a rational route that is most favored by the general public. On the other hand, when selecting a route other than the most rational route judged by the masses, users want to arrive as soon as possible, want to go cheap, want to stay seated, or want to leave without changing, etc. It is thought that there is a considerable influence. Therefore, by comparing the condition order of the first route with the condition order of the route actually selected by the user, it is possible to ask the tendency of the selection criterion by the user. To simplify the explanation, we will focus on two route conditions, fare and time, and explain the classification of user preferences. However, not only the combination of fare and time, but also add elements of route conditions such as the number of transfers. It is also possible to think. FIG. 14 is a diagram plotted on a plane based on the rank within the section of the fare and required time of the first route and the data of the fare of the route actually used by the user and the rank within the section of the required time. If the required time rank 801 is plotted on the vertical axis and the fare rank 802 is plotted on the horizontal axis, a user who often selects a route that is faster and cheaper than the first route collects points around the rational choice group 812. Will increase. In addition, users who often choose a route that can arrive early even if the fare is high gather points around the time priority group 813, and users who prefer a cheap route even if the time is late are around the fare priority group 811. Points will be collected. Furthermore, users who have a mix of fare priority cases and time priority cases where points are distributed around the different faction 814 are considered to have the property that priority elements are switched according to the situation. .

  FIG. 15 is a diagram showing a data structure for storing the result of estimating the behavioral preference regarding the movement of the user by the classification method as shown in FIGS. The selection criterion determination result list 380 includes a user ID 381, a target period 382 that is a determination target, a time zone 383 that is a determination target, the number 384 of movements of a plurality of route sections during the target period, and a plurality of times during the target period. Time priority point 385 calculated using connected movement data when moving a route section, fare priority point 386 calculated using connected movement data when moving a plurality of route sections during this target period, this target It includes information such as the number-of-transfers priority point 387 calculated using the connected movement data when moving over a plurality of route sections during the period, and the selection criteria 388 that is the result of determining the type of preference classification. To simplify the explanation here, we will focus on the three route conditions: fare, time, and number of transfers, and explain the data structure for storing the judgment results. In that case, an area for storing data of priority points of other elements may be prepared. In addition, for the period of data used for the determination, the determination result may be stored by classification of day of the week or weekday / holiday in addition to the target period and time zone.

FIG. 16 is a diagram showing a processing procedure for determining a behavioral preference when a user moves using a series of movement logs of individual users and storing it in the selection criterion determination result list 380. Here, in order to simplify the latter half of the processing procedure, focusing on two route conditions of fare and time, the data structure for storing the determination result will be described, but actually, other than the fare and time, Other route condition elements such as the number of transfers may be added. The following processing is repeated for the user ID (processing step 900). First, all connected travel logs corresponding to the preset target period and time zone and user ID are extracted, and the variable C, the time priority variable Pt, the fare priority variable Pf, and the total number of movements for multiple route sections are transferred. The number-of-times priority variable Pc is initialized (processing step 901). Here, it is assumed that the value of the target period and time zone is set in advance externally, and the same processing is performed for the combination of all the set data target time periods and time zones. An example of processing when the target period and time zone are determined is shown. The following processing is repeated for all the extracted post-connection movement logs (processing step 902). Referring to the departure station ID and arrival station ID after the connection travel log, station - on the basis of the area definition list, to calculate the arrival area ID as the departure area ID (processing step 903). Calculated departure area ID, the record of the section ID corresponding to the arrival area ID is checked whether there is to route order list 370, if it exists (processing step 904), the duration of the corresponding record, fare The rank within the section such as the number of transfers is extracted (processing step 905). Based on the extracted results, 1 is set for variable C that counts the number of section movements where multiple routes exist, 1 is the required time rank for time priority variable Pt, the fare rank is set for fare priority variable Pf, and the number of transfers priority variable. The rank of the number of transfers is added to Pc (processing step 906). Calculated departure area ID, the record of the section ID corresponding to arrival area ID does not exist on the path ordered list 370 (process step 907), the process proceeds to after the next connection travel log (processing step 908) . After all the connected movement logs have been determined, the time priority variable Pt, the fare priority variable Pf, and the transfer frequency priority variable Pc are divided by the movement frequency C to calculate an average value (processing step 909). Next, the process proceeds to the process of determining palatability using the average value, but here we will explain the two conditions of required time and fare in order to simplify the explanation. The present invention is not limited to these two conditions, and other conditions such as the shortest time and the number of transfers may be added. In order to determine which category the individual user belongs to, such as time priority or charge priority, identification is made by looking at the distribution of the priority variable distribution of each condition.

  First, the variable C indicating the number of movements is greater than 0, i.e., a user who has moved more than one route at least once, and the time priority variable Pt and the fare priority variable Pf remain at their initial values. In some cases (processing step 910), since all of the movements in the multi-path section will have moved the first route, it is generally considered a rational choice who prefers the most likely route, The result is stored in the selection criterion determination result list 380 (processing step 911). In addition, the variable C indicating the number of times of moving the multiple route section is greater than 0, that is, the user who has moved the multiple route section at least once, and the value of the time priority variable Pt is the fare priority variable Pf If the value is larger than the constant multiple of (Pt> Pf × D) (processing step 912), it can be regarded as a user who places more importance on time than fare. Store (processing step 913). On the other hand, the variable C indicating the number of times of moving the multiple route section is greater than 0, that is, the user who has moved the multiple route section at least once, and the value of the fare priority variable Pf is the time priority variable Pt If it is larger than the constant multiple of the value (Pf> Pt × D) (processing step 914), it can be regarded as a user who wants to travel on the cheapest route that emphasizes fare over time, so as a fare priority group The result is stored in the selection criterion determination result list 380 (processing step 915). Here, various methods can be considered for comparing the time priority variable Pt and the fare priority variable Pf in addition to the above, considering the average value, median value, and variance value. The variable C, which indicates the number of times the multi-path section was last moved, is greater than 0, that is, the user who has moved the multi-path section at least once, and the value of the fare priority variable Pf and the time priority variable Pt In the case of a user whose value is difficult to obtain (Processing Step 916), it is considered that the priority of the fare and the time changes according to the scene. The data is stored in 380 (processing step 917).

  FIG. 17 is a diagram illustrating an example of a presentation screen generated and distributed by the information distribution server 113 and a screen example of a user's preference-specific transfer guidance map. Station A1001, Station B1002, Station C1003, Station D1004, Station E1005, Station F1006, Station G1007 and route 1 (1011), route 2 (1012), route 3 (1013), route 4 (1014) are in the form of a schematic map. This is an example in which the route when traveling from the area of the station A 1001 to the area of the station B 1002 is presented according to preference. The route from the area of the station A 1001 to the area of the station B 1002 includes the route from the A station 1001 to the C station 1003 first on the route 1 (1011), transfer to the route 4 (1014) and getting off at the B station 1002, and the A station Go to station D1004 from E station 1005, which is a nearby station, by way of line 3 (1013), change to station 2 from station G, which is a nearby station, and get off at F station 1006, which is a neighboring station of station B1002. Suppose there are two patterns. When these two routes are routes that cost less fare but take longer, and routes that cost more but arrive early, users can easily understand the difference between the advantages and conditions of each route. It is possible to intuitively grasp and select a route that suits one's preference.

  FIG. 18 is an example of a presentation screen generated and distributed by the information distribution server 113, showing a screen example of user information directed to a provider around the station. A station K1031, a station L1032, and a station M1033 existing on the route 5 (1021) are displayed in a substantially map form, and the ratios are shown by classifying users of each station according to route selection criteria. . Here, the user of the station is, for example, an entry station ID, an entry station ID, a departure station ID, an arrival station ID, or the like of the movement log data 124 or the connected movement log data 125 when a certain target period or time zone is determined. Is a collection of user IDs of records that include the code of the target station. The user distribution of each station can be created by estimating the selection criteria at the time of movement for the collected user IDs according to the above-described processing procedure, and performing aggregation for each category. For example, a station with many time-priority users will be provided with a store that can quickly shop, or a station with many fare-priority users will receive a low price. It can be expected to be used in marketing areas such as advertising

Information for generating the presentation screens of FIGS. 17 and 18 is accumulated in the storage unit 137 of the calculation server 112, and the users 115 and 117 access a predetermined Web page and select a station or area by using a pull-down menu or the like. The search processing program 141 is executed according to a condition specified by selection or the like, and necessary information is acquired. The information acquired by the information distribution program 142 is edited and information is distributed.
FIG. 19 is a diagram showing an example of a transfer route search screen 1100 for users generated and distributed by the information distribution server 113. The search processing program 141 is executed in response to the conditions specified by the users 115 and 117 on the presented transfer route search screen 1100, and the result information is edited and distributed by the information distribution program 142. First, the users 115 and 117 use the information terminals 116 and 118 to select display contents from a pull-down menu on a Web screen, for example. These display conditions can be set and changed by the users 115 and 117 using an input interface such as a setting screen or a mouse / keyboard. For example, on the transfer route search screen 1100 for the users 115 and 117, the member ID 1001, the departure station 1102, the arrival station 1103, the date 1104, the time 1105, the type 1106 for selecting whether the departure time or the arrival time, and the search result display order 1107 Is selected from a pull-down menu or the like, and the designated condition is passed to the search processing program 141. As for the input of the member ID 1001, the ID of the user's transportation IC card may be transmitted to the information distribution server 113 through the IC card reader 119. Further, the member ID 1001 may be stored in the information distribution server 113 when the users 115 and 117 are accessed for the first time, and the input may be omitted after the next time. At this time, it is assumed that the menu such as date and time is preset with the current date and time, and the user does not necessarily have to select it. In general, the display order on the transfer route search screen is determined according to the rules of the transfer guide engine provider, and users of the transfer guide engine must enter each one, such as whether to give priority to fare or time. Although it has been necessary to use the personal selection criterion estimation result during movement according to the present invention, it is possible to automatically determine a route that matches the searcher's preference and present it preferentially. Even in sections that are used for the first time, it is possible to estimate the preference in other usage sections from the past movement history with fine granularity such as weekdays and holidays, and to determine the route that seems to be most appropriate for that person it can. Moreover, it is also possible to learn a preference time-series change by using a long-term history.

  FIG. 20 is a diagram illustrating an example of a transfer route search result screen 1200 for a user generated and distributed by the information distribution server 113. The search processing program 141 is executed in accordance with the menu selected on the transfer route search screen 1100 by the user, and the result information is edited and distributed by the distribution program 142. In the transfer route search result screen 1200 for users, a screen in which the search results are rearranged for the fare priority group is presented as an example. As a result of executing the search processing program 141 for the combination of the departure point A station and the arrival station B station, a pattern for changing at the C station and a pattern for changing at the D station are extracted. The screens are arranged in the order that suits the searcher's preference, and the pattern that changes at station C is route 1 and the pattern that changes at station D is route 2 and is distributed to the user.

  In the above example, the usage history of the traffic non-contact IC card is used to obtain the behavioral preference when the individual moves. However, in addition to this, the location information acquired by the mobile phone or the purchase It is possible to improve accuracy and quality by analyzing usage history such as information together.

  As described above, by analyzing the behavior data of station users, the behavioral preference obtained according to each user's actual usage is estimated, and the information is processed appropriately according to the user's request. Can be provided.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made. It is possible to appropriately combine the above-described embodiments. It will be understood by the contractor.

001 to 007 ... station, 011 to 014 ... route, 101 ... user (IC card owner), 102 ... ticket gate, 103 ... IC card, 104 ... network, 105 ... data server, 111 ... history server, 112 ... calculation Server 113 113 Information distribution server 114 Internet 115 User 119 Mobile information terminal 117 User 118 Information terminal 119 IC card reader 121 Data storage unit 122 Log with ID Data, 123 ... Master data, 124 ... Movement log data, 125 ... Connected movement log data, 131 ... CPU, 132 ... Memory, 133 ... Storage unit, 134 ... Movement log generation program, 135 ... Movement log connection program, 136 ... Representative route calculation program, 137... Action preference determination program, 138. , 139 ... Network interface, 141 ... Search processing program, 142 ... Information distribution program, 143 ... Network interface, 144 ... Network interface, 145 ... CPU, 146 ... Memory, 147 ... Storage unit, 241 ... Log ID, 242 ... User ID , 243 ... Station ID, 244 ... Usage time, 245 ... Usage type, 260 ... Station data master, 261 ... Station ID, 262 ... Station name, 263 ... Railway company, 264 ... Location, 265 ... Latitude / longitude, 270 ... Route master , 271 ... route ID, 272 ... route name, 273 ... railway company, 274 ... route type, 280 ... station-route relation master, 281 ... route ID, 282 ... station ID, 283 ... order, 284 ... station type, 285 ... Time required from the first station, 290 ... route master, 291 ... route D, 292 ... Entrance station ID, 293 ... Participating station ID, 294 ... Period, 295 ... Time zone, 296 ... Route ID 1,297 ... Transfer station ID 1,298 ... Route ID 2,301 ... Number of boarding times, 302 ... Standard time, 303 ... shortest time, 304 ... fee, 321 ... log ID, 322 ... user ID, 323 ... entry date and time, 324 ... entry date and time, 325 ... entry station, 326 ... entry station, 327 ... payment amount, 328 ... route ID, 329 ... Purchase flag, 331 ... connected log ID, 332 ... user ID, 333 ... departure date and time, 334 ... arrival date and time, 335 ... departure station ID, 336 ... arrival station ID, 337 ... movement log ID 1,338 ... movement log ID 2,340 ... Station-area definition list, 341 ... Area ID, 342 ... Representative station ID, 343 ... Target period, 344 ... Number of stations, 345 ... Station ID1, 346 ... Station D2,347 ... station ID 3,348 ... station ID 4,323 ... entrance date and time, 350 ... section-route list, 351 ... section ID, 352 ... departure area ID, 353 ... arrival area ID, 354 ... target period, 355 ... Time zone, 356 ... Number of routes 357 ... Route ID 1, 358 ... Route ID 2, 370 ... Route ranking list, 371 ... Section ID, 372 ... Departure area ID, 373 ... Arrival area ID, 374 ... Target period, 375 ... Time zone, 376... Route, 377. Order within required time section, 378. Order within fare section, 379 ... Order within transfer section, 400 to 411 ... Processing step, 500 to 509 ... Processing step, 600 to 606 ... Processing step, 700 to 706 ... Processing step, 801 ... Axis of required time rank (direction slower than the first route), 802 ... Axis of fare rank (first route) 811 ... Fare priority, 812 ... Rational choice, 813 ... Time priority, 814 ... Different use, 380 ... Selection criteria judgment result list, 381 ... User ID, 382 ... Target period, 383 ... Time zone , 384 ... Number of times of using a plurality of route sections, 385 ... Time priority point, 386 ... Fare priority point, 387 ... Transfer frequency priority point, 388 ... Selection criteria, 900 to 917 ... Processing steps, 1001 to 1007 ... Station, 1011 1014 ... route, 1021 ... route, 1031 to 1033 ... station, 1100 ... route search screen, 1101 ... user ID, 1102 ... departure station, 1103 ... arrival station, 1104 ... date, 1105 ... time, 1106 ... type, 1107 ... display Order, 1200 ... Search result screen.

Claims (8)

Use type information indicating admission and participation, log ID, user ID, station ID and log data with ID including station time, route ID, admission station ID, exit station ID, fare information, required time, and number of transfers A data storage unit for storing a route master, a station area definition list including an area ID, a representative station ID, and a plurality of station IDs;
Based on said ID-log data stored in the data storage unit and the route master, the log ID and the user ID and entry date and played date the entrance station ID and the participating station ID and the route ID A movement log generation unit that generates a plurality of movement logs including,
Based on the admission date and time and the participation date and time and a predetermined connection determination threshold of the plurality of generated movement logs, a movement log connection unit that connects the plurality of movement logs,
Based on the route master and the station area definition list stored in the data storage unit, an interval route list including an interval ID, a departure area ID, an arrival area ID, the number of routes, and the route ID is generated. ,
Based on the number of routes in the section route list and the route master, the required time, the fare information, and the number of transfers are calculated for the section ID, and the required time rank, the fare information rank, and the number of transfers of the route ID are calculated. and the calculated ranking, a representative route calculating unit for generating a route ranking list including said segment ID wherein the departure area ID the arrival area ID and the required time order and said fare information order and the transfer count ranking,
Based on the generated route order list, a required time priority, a fare information priority, and a transfer frequency priority are generated for each user ID, and the user ID, the required time priority, and the fare information priority, An action preference determination unit that generates a selection criterion determination result list including the transfer frequency priority;
Accepting user input, based on the selection criterion determination result list, extracts the required time priority and fare information priority and the number of times of transfer priority corresponding to the input of the accepted user, the required issued extracted An interface unit for displaying output information on a screen based on time priority, the fare information priority, and the transfer frequency priority;
A personal behavior preference analysis system characterized by comprising: The personal behavior preference analysis system according to claim 1,
The user input includes a first user ID, a first station ID, and a second station ID,
The output information includes a route connecting the first station ID and the second station ID, a first required time priority corresponding to the first user ID, and a first fare information priority. The personal action preference analysis system characterized by including the information based on the 1st transfer frequency priority. The personal behavior preference analysis system according to claim 1,
The user input includes a first station ID;
The output information includes information based on a first required time priority, a first fare information priority, and a first transfer frequency priority corresponding to a user ID having a record of the first station ID. Personal behavior preference analysis system. The personal behavior preference analysis system according to claim 2,
The user input includes priority display order information indicating a priority when displaying the required time priority, the fare information priority, and the transfer frequency priority,
The interface unit displays the output information on the screen based on the priority display order information, and the personal behavior preference analysis system. Use type information indicating admission and participation, log ID, user ID, station ID and log data with ID including station time, route ID, admission station ID, exit station ID, fare information, required time, and number of transfers A personal behavior preference analysis method for analyzing personal behavior preferences using a data storage unit that stores a route master, a station area definition list including an area ID, a representative station ID, and a plurality of station IDs,
Based on said ID-log data stored in the data storage unit and the route master, the log ID and the user ID and entry date and played date the entrance station ID and the participating station ID and the route ID A step of generating a plurality of movement logs including a movement log generation unit;
Connecting the plurality of movement logs by a movement log connection unit based on the entry date and time and the participation date and time and a predetermined connection determination threshold of the plurality of generated movement logs;
Based on the route master and the station area definition list stored in the data storage unit, a section route list including a section ID, a departure area ID, an arrival area ID, the number of routes, and the route ID is represented as a representative route. Generated by the calculator,
Based on the number of routes in the section route list and the route master, the required time, the fare information, and the number of transfers are calculated for the section ID, and the required time rank, the fare information rank, and the number of transfers of the route ID are calculated. calculated with ranks, generates a route ranking list including said segment ID wherein the departure area ID the arrival area ID and the required time order and said fare information rank and the number of times of transfer order by the representative route calculator Process,
Based on the generated route order list, a required time priority, a fare information priority, and a transfer frequency priority are generated for each user ID, and the user ID, the required time priority, and the fare information priority, A step of generating a selection criterion determination result list including the transfer frequency priority by an action preference determination unit;
Accepting user input, based on the selection criterion determination result list, extracts the required time priority and fare information priority and the number of times of transfer priority corresponding to the input of the accepted user, the required issued extracted Displaying output information on the screen by the interface unit based on the time priority, the fare information priority, and the transfer frequency priority;
A personal behavior preference analysis method characterized by including: The personal behavior preference analysis method according to claim 5,
The user input includes a first user ID, a first station ID, and a second station ID,
The output information includes a route connecting the first station ID and the second station ID, a first required time priority corresponding to the first user ID, and a first fare information priority. The personal action preference analysis method characterized by including the information based on the 1st transfer frequency priority. The personal behavior preference analysis method according to claim 5,
The user input includes a first station ID;
The output information includes information based on a first required time priority, a first fare information priority, and a first transfer frequency priority corresponding to a user ID having a record of the first station ID. The personal behavior preference analysis method characterized by this. The personal behavior preference analysis method according to claim 6,
The user input includes priority display order information indicating a priority when displaying the required time priority, the fare information priority, and the transfer frequency priority,
In the step of displaying the output information on the screen by the interface unit, the output information is displayed on the screen based on the priority display order information.

Images