JP5419854B2 - Action base identification system, action base identification method and action base identification program - Google Patents

Description

  The present invention relates to information processing technology using GPS (Global Positioning System) and the like, and more particularly, to technology for processing positioning information of a mobile terminal positioned using GPS.

  Various services using GPS are provided. As an example of such a service, the following Patent Document 1 includes means for setting information related to electronic mail and a transmission point on a route, means for calculating the current vehicle position of the vehicle, and the calculated vehicle position. A navigation device having means for monitoring whether or not the set transmission point has been reached and transmission means for transmitting the set e-mail when the vehicle position has reached the transmission point is disclosed. Yes.

JP 2000-285382 A

  By the way, with the rapid spread of mobile terminals, the Internet, navigation systems, etc., the use of GPS is expanding in various scenes. As a result, there is an increasing need for improving the convenience of services using GPS among users of mobile terminals equipped with a GPS function. For example, as an example of a service that uses the GPS function of a mobile terminal, when the current position of the mobile terminal is periodically measured and the current position that has been measured approaches a destination set in advance by the user, that fact is indicated. There are known services for notifying a user and providing information related to the destination.

  Here, it is conceivable that related information is automatically transmitted based on the current position of the user for a predetermined facility or place that is not set by the user. However, in this case, all users in the vicinity of the target area are the transmission targets of the related information. Depending on the relationship between the target area and the user's action base, the related information may be useful or unnecessary. Sometimes it becomes. For example, when the target site is a tourist destination, sending tourist information to a user who has visited for sightseeing increases convenience for the user, while providing tourist information to a user living in the tourist destination. Transmission of the user's convenience is adversely affected. When the target site is a shopping mall, the weekend sale information of the shopping mall is useful for a user who has a home in the vicinity of the target site, but is likely not necessary for a user who happens to visit.

  Therefore, in order to avoid such a problem, a method in which the user's own action base (for example, home, work place, school, favorite shop, shopping center, golf course, etc.) is set in advance by the user can be considered. . However, requesting such setting work from the user lacks convenience of the service. In addition, since the action base can change, the user must change the setting every time the action base changes, and the work tends to be complicated.

  Therefore, this invention is providing the action base identification apparatus which can specify a user's action base automatically.

  The behavior base identifying system includes a storage unit that stores a positioning point obtained by positioning the current position of the mobile terminal transmitted from the mobile terminal and a positioning time of the positioning point in a predetermined database, and the predetermined database. A basis for generating basic data in which the positioning point and the positioning time stored in a map are associated with a mesh region to which the positioning point belongs among a plurality of mesh regions obtained by dividing a map into a mesh shape Based on the data generation means and the generated basic data, the number of positioning days in which the positioning point is positioned in each mesh area within a predetermined period, the number of times the positioning point is positioned, and the user of the mobile terminal stay Calculate at least two of the staying hours as action base parameters, and associate each mesh region with the calculated two action base parameter values. Mesh data generating means for generating mesh data, and using the generated two action base parameter values according to respective priorities, action base likelihood calculating means for calculating the action base likelihood of the mesh region; Action base specifying means for specifying the mesh area having the largest likelihood of action base as the action base mesh area including the user's action base from among the plurality of mesh areas. .

  Further, out of the plurality of mesh areas, a daily area setting means for setting, as the daily area of the user, a mesh area in which the value of the generated two action base parameters is equal to or greater than a predetermined daily area threshold; Group processing means for grouping mesh regions satisfying a predetermined connection relationship for mesh regions included in the set daily sphere, and calculating group-specific action base parameters for each grouped mesh group, The behavior base likelihood calculation means identifies a behavior base mesh group from the mesh group based on the calculated group-specific behavior base parameter, and the behavior base for each mesh region in the identified behavior base mesh group The behavior base identifying means calculates the likelihood, and the behavior base identification means identifies the identified behavior base message. From each mesh region in Interview group, and specifying the calculated basing likeness largest mesh regions as basing mesh region of the user.

  Further, when the user's action base is the first action base that is the home of the user's life, the two action base parameters include the number of positioning days having the first priority and the stay having the second priority. Time is set, and the behavior base likelihood calculation means satisfies the predetermined first behavior base calculation condition and the maximum number of positioning days by group from the mesh group. The first action base mesh group is specified as the first action base mesh group, and the first action base mesh group in the specified first action base mesh group is used according to the priority order for the first action base time and the staying time. A value indicating the likelihood of the site is calculated.

  Further, when the user's action base is the second action base where the user stays continuously, the two action base parameters include a stay time having a first priority and a positioning day having a second priority. The behavior base-likeness calculating means is configured to select, from the mesh group, a mesh having the maximum number of days for positioning by group and a stay time by group satisfying a predetermined second action base calculation condition and having the maximum stay time by group By identifying the group as a second action base mesh group and using the stay time and the number of positioning days according to each priority for each mesh region in the specified second action base mesh group, It is characterized in that a value indicating is calculated.

  In addition, when a plurality of action bases are specified, the action base specifying means matches the action base mesh area of one specified action base with the action base mesh area of another action base specified in the past. If so, the behavior base likelihood calculation means is controlled to recalculate the behavior base likelihood of the one behavior base for a mesh group excluding the mesh group including the behavior base mesh region of the other behavior base. It is characterized by doing.

  Further, when a plurality of action bases are specified, the action base likelihood determination unit can specify a mesh group of one action base, and if the mesh group of another action base cannot be specified, The behavior base likelihood of the other behavior base is calculated for a mesh region other than the behavior base mesh region of the one behavior base included in the mesh group of the behavior base.

  In addition, the mesh data generation means is configured to determine the second positioning point from the first positioning time of the first positioning point for the second positioning point arranged next to the first positioning point and the first positioning point on the time axis. Calculating a time between positioning points up to the second positioning time and a distance between positioning points between the second positioning point and the first positioning point, and based on the time between the positioning points and the distance between the positioning points. In accordance with the determined distribution condition, a distribution means for distributing the time between the positioning points as a staying time to a mesh region on the connection connecting the first positioning point and the second positioning point is provided, and the allocated time is Based on this, the stay time in the mesh region is calculated.

  In addition, when the time between the positioning points is less than a predetermined time and / or the distance between the positioning points is less than the predetermined distance, the distribution means determines the second positioning point from the first positioning point based on the time between the positioning points. The movement time subtracted time obtained by subtracting the movement time to is distributed to the first mesh area to which the first positioning point belongs, and the movement time is substantially evenly distributed to the mesh area on the connection. And

  In addition, when the time between the positioning points is less than a predetermined time and / or when the distance between the positioning points is not less than the predetermined distance, the distribution means determines the second positioning point from the first positioning point based on the time between the positioning points. The movement time subtracted time obtained by subtracting the movement time to is distributed to the first mesh area to which the first positioning point belongs, and the movement time is not distributed to the mesh area on the connection.

  In addition, when the first positioning time of the first positioning point and the second positioning time of the second positioning point span a predetermined reference time, the distribution unit is configured to obtain the predetermined positioning time from the first positioning time to the predetermined reference time. The maximum time is allocated to the first mesh area, and the time from the second reference time to the predetermined reference time is allocated to the second mesh area as a maximum.

  In addition, the action base identifying system of the present invention includes a storage unit that stores a positioning point obtained by positioning the current position of the mobile terminal transmitted from the mobile terminal and a positioning time of the positioning point in a predetermined database in association with each other. Basic data in which the positioning point and the positioning time stored in the predetermined database are associated with a mesh region to which the positioning point belongs among a plurality of mesh regions obtained by dividing a map into a mesh. Basic data generating means for generating the positioning data, the number of positioning days in which each positioning point is positioned in each mesh region based on the generated basic data, the number of positioning times when the positioning point is positioned, and the portable terminal At least two of the stay times during which the user stayed are calculated as action base parameters, and each mesh region and the calculated two action base parameters are calculated. Mesh data generating means for generating mesh data in association with each other, and a mesh having the largest value of at least one action base parameter in accordance with the priority order among the generated two action base parameters from among the plurality of mesh regions Action base specifying means for specifying an area as an action base mesh area including the user's action base.

  Moreover, this invention is materialized also as an action base identification method. The behavior base identification method includes a step of storing in a predetermined database a positioning point obtained by positioning the current position of the mobile terminal transmitted from the mobile terminal and a positioning time of the positioning point, and storing the positioning point in the predetermined database Generating basic data in which the positioning point and the positioning time are associated with a mesh region to which the positioning point belongs among a plurality of mesh regions obtained by dividing a map into a mesh shape; Based on the generated basic data, among the number of positioning days in which the positioning point is positioned in each mesh area within a predetermined period, the number of times the positioning point is positioned, and the staying time during which the user of the mobile terminal stayed Mesh data in which at least two of these are calculated as action base parameters, and each mesh region is associated with the calculated two action base parameter values Generating the behavior base likelihood of the mesh region by using the generated two behavior base parameter values according to the respective priorities, and from among the plurality of mesh regions, Identifying the mesh area having the highest likelihood of an action base as the action base mesh area including the user's action base.

  Further, the present invention is also realized as a program for causing a computer to execute the above steps. This program can be installed or loaded on a computer through various recording media such as an optical disk such as a CD-ROM, a magnetic disk, or a semiconductor memory, or via a communication network.

  Further, in this specification and the like, the term “means” does not simply mean a physical means, but includes a case where the functions of the means are realized by software. Further, the function of one means may be realized by two or more physical means, or the functions of two or more means may be realized by one physical means.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the action base identification apparatus which can specify a user's action base automatically based on the said user's movement.

It is a block diagram which shows schematic structure of an action base identification system. It is a figure explaining an example of basic data. It is a figure explaining an example of daily mesh data. It is a figure explaining the time allocation between positioning points. It is a figure explaining an example of the mesh data according to a period. It is a figure explaining the setting of an everyday zone. It is a figure which shows an example of a mesh group. It is a figure explaining an example of an action base mesh group and an action base mesh. It is a figure which shows an example of the data structure of positioning information DB, basic data DB, and daily mesh DB. It is a figure which shows an example of the data structure of mesh DB according to a period, and daily life DB. It is a figure which shows an example of the data structure of action base DB. It is a block diagram which shows schematic structure of a portable terminal. It is a flowchart which shows an example of the whole flow of an everyday sphere setting process. It is a flowchart which shows an example of the flow of a daily mesh data generation process. It is a flowchart which shows an example of the flow of a mesh data generation process according to a period, and a daily sphere setting process. It is a flowchart which shows an example of the whole flow of an action base specific process. It is a flowchart which shows an example of the whole flow of a home determination process. It is a flowchart which shows an example of the whole flow of work place determination processing. It is a figure for demonstrating a home determination confirmation and a work location determination confirmation. It is a flowchart which shows an example of the whole flow of a home determination confirmation process. It is a flowchart which shows an example of the flow of the whole work location determination confirmation process.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The action base identification system of the present embodiment automatically sets a user's daily life area (hereinafter referred to as “daily area”) using positioning information from the terminal, and from the set daily areas, Identify a user's specific location of action. As a method for identifying the user's action base from the daily life area, for example, it may be possible to use the number of times of positioning or the like as the action base parameter. However, if only one type of action base parameter is used, the accuracy of specifying the action base is poor. There is a problem. Therefore, in the present invention, the action base is specified by using at least two or more kinds of action base parameters in accordance with the priority order, thereby improving the accuracy of the action base specification. Hereinafter, in the present embodiment, the first action base (for example, “home”) that is assumed to have the strongest connection with the user and the base of the life of the user who stays continuously is the base of the user's life. The case where the 2nd action base (for example, "work location" and "school") with which the connection with the place is not so strong is each specified is demonstrated.

[1. Configuration of action base identification system]
FIG. 1 is a block diagram showing an example of a schematic configuration of the action base identifying system according to the present embodiment. The action base specifying system (action base specifying apparatus) 10 is connected to a mobile mobile terminal device (hereinafter referred to as “mobile terminal”) 20 via a network N.

  The action base identification system 10 includes a main control unit 101, a communication unit 102, a positioning information storage unit 103, a basic data generation unit 104, a mesh data generation unit 105 (daily mesh generation unit 106 (stay seconds / position count counting unit 107). Distribution complementing means 108), period-specific mesh generating means 109), daily sphere setting means 110, mesh group processing means 111, action base likelihood calculation means 112, action base specifying means 113, positioning information DB 114, basic data DB 115, It mainly includes various function realizing means such as a daily mesh DB 116, a period-specific mesh DB 117, an everyday area DB 118, and an action base DB 119.

  Although not shown in the figure, the action base specifying system 10 includes a CPU, a ROM, a RAM, and other memories, an external storage device for storing various information, an input / output interface, a communication interface, and a bus connecting them. For example, when the CPU executes a predetermined program stored in a ROM or the like, it functions as each of the function realizing means. In addition, even if the action base identification system 10 is comprised from a single computer, it may be comprised from the some computer distributed on the network.

  The main control unit 101 controls the entire operation of the action base specifying system 10 and the operation of each unit described above. The communication means 102 is an interface for transmitting and receiving various types of information by communicating with the mobile terminal 20 via the network N. The communication unit 102 also functions as a reception unit that receives positioning information transmitted from the mobile terminal 20.

  The positioning information storage means 103 stores the positioning information transmitted from the portable terminal 20 in the positioning information DB 114. Although the positioning information will be described later, a positioning point (coordinate information) where the current position of the mobile terminal 20 is positioned, a positioning time of the positioning point, and user identification information (ID) that uniquely identifies the user of the mobile terminal 20 And are included.

  Based on the positioning information stored in the positioning information DB 114, the basic data generating means 104 generates basic data in which positioning points and positioning times are associated with corresponding mesh regions of the map for each user and for each positioning date. The mesh area is a plurality of divided areas obtained by dividing the map into meshes based on latitude and longitude, and the size of the divided areas can be set as appropriate according to the specifications and design (for example, the user If you want to specify an area where you often stay, you can set the size larger, while if you want to identify stores that users frequently visit, you can set the size smaller.) The corresponding mesh area means the mesh area to which the coordinate information of the positioning point belongs. FIG. 2 is a diagram for explaining how basic data is generated based on the positioning information stored in the positioning information DB 114. In the same figure, regarding the positioning date (2010/03/03) of the user (Mr. A), basic data B1 in which a positioning point and a positioning time are associated with each mesh area is described as a user's action trajectory on the corresponding day. Yes. The generated basic data is stored in the basic data DB 115.

  Returning to FIG. 1, the mesh data generation means 105 is based on the basic data stored in the basic data DB 115, and at least two of the number of positioning days, the number of positioning times, and the user's stay time within a predetermined period in each mesh region Is calculated as an action base parameter, and mesh data in which each mesh region is associated with the calculated at least two action base parameters is generated. The number of positioning days is the total number of days that positioning was performed in the corresponding mesh area within a predetermined period (the same day is not double-added) and indicates the number of days the user stayed in the corresponding mesh area. Is called "stay days". The number of positioning times is the total number of times positioning has been performed in the corresponding mesh region within a predetermined period. The user's stay time is the total time that the user stayed in the corresponding mesh area within a predetermined period. In the present embodiment, since the stay time is expressed in seconds, this is also referred to as “stay seconds” hereinafter. In the action base parameter, any combination of at least two of the stay days, the number of positioning times, and the stay seconds can be set according to the content of the action base to be specified. You may add other information. The mesh data includes daily mesh data and period-specific mesh data, which will be described below.

  The daily mesh generation means 106 generates mesh data for each unit period as data representing the number of times of positioning and the number of stay seconds in each mesh region for each predetermined unit period. The predetermined unit period is a minimum unit period for generating mesh data, and can be appropriately set according to specifications / designs such as “one day” and “half day”. In the present embodiment, “1 day” is set as the predetermined unit period, and hence the mesh data by unit period is hereinafter referred to as “daily mesh data”. FIG. 3 is a diagram for explaining how daily mesh data is generated based on basic data stored in the basic data DB 115. In the figure, mesh data M1 in which the number of stay seconds and the number of times of positioning are associated with each mesh area is described for the positioning date (2010/03/03) of the user (Mr. A). In this embodiment, the case where the number of positioning times and the number of staying seconds are calculated and associated for each mesh region has been described. However, only the number of positioning times or the number of staying seconds may be calculated and associated. The generated daily mesh data is stored in the daily mesh DB 116.

  Returning to FIG. 1, the daily mesh generation unit 106 specifically includes a staying seconds / positioning count totaling unit 107 and a distribution complementing unit 108.

  Based on the basic data stored in the basic data DB 115, the staying seconds / positioning number counting means 107 calculates the number of positioning points at each mesh area and the number of staying seconds of the user of the mobile terminal for each predetermined unit period. Calculate and associate each mesh region with the calculated number of positioning times and staying seconds. The number of positioning times in each mesh area is calculated by summing the number of times positioning has been performed in the corresponding mesh area. The staying seconds in each mesh area is calculated by totaling the allocation time allocated to the corresponding mesh area by the distribution complementing means 108 described later.

  The distribution complementing means 108 selects the first positioning point and the second positioning point arranged next to the first positioning point on the time axis from the positioning points associated with the mesh region. The first positioning point is also called a pre-movement point, and the second positioning point is also called a post-movement point. Next, the time from the first positioning time of the first positioning point to the second positioning time of the second positioning point (hereinafter referred to as “time between positioning points”), the second positioning point, and the first positioning point (Hereinafter, referred to as “distance between positioning points”). Then, an allocation condition is determined based on the time between the positioning points and the distance between the positioning points, and the time between the positioning points is set on the movement path from the first positioning point to the second positioning point according to the determined allocation condition. Allocate to the mesh area (including the first mesh area of the first positioning point and the second mesh area of the second positioning point). The allocation includes allocation of 0 hours. In this case, the allocation is substantially the same as not allocation.

  The distribution conditions can be set as appropriate according to the specifications and design, and there is no particular limitation on the contents thereof, but in this embodiment, according to the transmission conditions and configuration of positioning information from the mobile terminal, as follows: Allocation conditions are set. Specifically, the mobile terminal 20 according to the present embodiment transmits positioning information to the behavior base identifying system 10 at a predetermined transmission time interval (for example, 5 minutes). The positioning information is not transmitted for the reason. That is, when the user is in the radio wave range, the positioning information from the mobile terminal 20 is transmitted at a predetermined transmission time interval only when the user is moving. Therefore, the user's stay at the first positioning point, which is the point before moving, is the time obtained by excluding the moving time from the time between the positioning points between the first positioning time and the second positioning time (hereinafter referred to as “time after moving time subtraction”). Therefore, the movement time subtracted time is allocated to the mesh area of the first positioning point.

  Next, when it is possible to estimate the user's movement content (moving means and movement route) between the first positioning point and the second positioning point (when there is continuity between the first positioning point and the second positioning point) ) Distributes the movement time substantially evenly to the mesh region located on the estimated movement path (hereinafter referred to as “estimated movement path”) from the first positioning point to the second positioning point (first distribution). conditions). That is, the staying seconds are complemented for the mesh region on the estimated movement route. On the other hand, when the user's movement content between the first positioning point and the second positioning point cannot be estimated (when there is no continuity between the first positioning point and the second positioning point), the mesh area of the movement time No allocation is made (0 hours are allocated) (second allocation condition). That is, the number of seconds staying in the mesh area is not complemented.

  The method for determining whether or not the movement content can be estimated can be appropriately set according to the specification and design, and the content thereof is not particularly limited. For example, the time between positioning points and / or the distance between positioning points is a predetermined reference time. If it is less than the distance and / or less than the predetermined reference distance, it is assumed that the movement path of the user can be estimated by assuming that the movement is by walking or the like. On the other hand, when it is not less than the predetermined reference time and / or less than the predetermined reference distance, it can be assumed that the user's movement route cannot be estimated by considering that the movement is, for example, when using the subway.

  In addition, the calculation method of the travel time can be appropriately set according to the specification and design, and the content thereof is not particularly limited. For example, a method of using a transmission time interval of the mobile terminal 20 or a distance between positioning points And a method of calculating the travel time from a predetermined speed according to the travel means (eg, train, car, walk). In the present embodiment, in the case of the first distribution condition, the transmission time interval (for example, 5 minutes) of the positioning information in the mobile terminal 20 is set as the travel time. In the case of the second distribution condition, the travel time from the first positioning point to the second positioning point is calculated based on the distance between the positioning points and a predetermined speed (for example, 30 km) set in advance. Note that the predetermined hourly speed can be appropriately set according to the estimated moving means and the like. For example, for the first distribution condition, the traveling time may be calculated using the hourly walking speed.

  Also, the method for specifying the movement route can be set as appropriate according to the specification and design, and the content thereof is not particularly limited. In the present embodiment, however, the connection is made by connecting the first positioning point and the second positioning point. Represents the travel route. In FIG. 4A, the connection is shown as a straight line, but the connection is not limited to a straight line, and has, for example, a shape along an arbitrary movement route estimated according to a road, a route, a building, or the like. be able to. Further, among the mesh regions located on the connection, mesh regions other than the mesh regions of the first positioning point and the second positioning point are referred to as complementary mesh regions.

  In addition, since the daily mesh generation means 106 manages the positioning information for each predetermined unit period, when the first positioning point and the second positioning point cross the day (for example, the reference time (e.g., 0: 00/24 When)). That is, the first positioning point (hereinafter referred to as “first positioning point”) and the last positioning point (hereinafter referred to as “last positioning point”) in the unit period before the movement to be combined. A positioning point or a post-movement positioning point belongs to another day. Therefore, in the present embodiment, as the third distribution condition, for the first positioning point, since the corresponding pre-movement positioning point belongs to another day, the time from the reference time (for example, 0:00) to the positioning time of the first positioning point. The time is allocated as the staying seconds to the mesh area of the first positioning point. For the last positioning point, since the corresponding post-movement positioning point belongs to another day, the time from the positioning time of the last positioning point to the reference time (for example, 24:00) is displayed in the mesh area of the last positioning point. To distribute. In the third distribution condition, the maximum time allocated to the mesh area can be set, and the content thereof is not particularly limited, but in the present embodiment, a maximum of 6 hours is set.

  FIG. 4A is a diagram illustrating a state in which the staying seconds are distributed according to the first distribution condition. In the figure, first, the travel time obtained by subtracting the travel time (5 minutes) from the time between the positioning points between the first positioning point and the second positioning point (9: 40-9: 30 = 10 minutes). The subtracted time (5 minutes (300 seconds)) is distributed to the mesh area of the first positioning point. Next, in order to supplement the stay time in the mesh area between the first positioning point and the second positioning point, the movement time (5) is set to the six mesh areas located on the connection between the first positioning point and the second positioning point. Minute) is distributed approximately evenly (50 seconds each).

  FIG. 4B is a diagram for explaining a state in which the staying seconds are distributed according to the second distribution condition. In the figure, first, the travel time (20 minutes) between the first positioning point and the second positioning point is calculated from the distance between positioning points (10 km) and a predetermined speed (eg, 30 km / hour). The movement time subtracted time (100 minutes) obtained by subtracting the movement time (20 minutes) from the interval time (11: 30-9: 30 = 120 minutes) is allocated to the mesh area of the first positioning point. Yes. Here, since the mesh region located between the first positioning point and the second positioning point is not complemented, the travel time is not allocated.

  FIG. 4C is a diagram illustrating a state in which the staying seconds are distributed according to the third distribution condition. In the figure, the first positioning point (positioning time 7:30) is allocated 6 hours as the maximum time, and the last positioning point (positioning time 23:30) is from the positioning time to 0:00. 30 minutes are allocated.

  Next, returning to FIG. 1, the period-specific mesh generation means 109 will be described. The mesh generator 109 for each period totals the number of positionings and the number of stays included in the daily mesh data for each predetermined counting period, and associates the number of times of positioning, the number of stays, and the number of stays with each mesh area. Total period-specific mesh data (hereinafter referred to as “period-specific mesh data”) is generated. The predetermined counting period can be set as appropriate according to the design and specifications, and the contents of the period are not particularly limited. For example, the counting period such as the latest 7 days, the latest 30 days, the latest 180 months, the latest 365 days, etc. Can be set. A plurality of total periods may be set at the same time. In addition, if mesh data by period already exists, the daily mesh data for the date to be newly added (for example, the latest positioning date) and the date to be subtracted (for example, the oldest positioning date in the existing counting period) Read from the daily mesh DB 116 and add the stay seconds, positioning times, stay days for the additional target day for each mesh area, while subtracting stay seconds, positioning times, stay days for the subtraction target day for each mesh area To do. FIG. 5 is a diagram for explaining how the period-specific mesh data is generated from the daily mesh data. This figure shows a state in which the number of stay seconds, the number of positioning times, and the number of stay days are associated with each mesh area in the mesh data M6 by period generated from the daily mesh data M5.

  The daily area setting unit 110 extracts a mesh region in which at least two combinations of the number of positioning times, the number of staying seconds, and the number of staying included in the period-specific mesh data are equal to or greater than a predetermined threshold, Set the daily life zone.

  FIG. 6 is a diagram for explaining the setting of the daily sphere based on the period-specific mesh data. In the figure, when a mesh area that matches the condition of “stay time is 300 minutes or more and stay days is 3 days or more” is extracted from the mesh data M7 of the most recent 30 days, A case is shown in which the daily sphere is set by extracting a mesh area that matches the condition “stay time is 60 minutes or more and stay days are 2 days or more” from the mesh data M8 of the day. As shown in the figure, the daily life area may include various places such as a user's main base of action, home, work place, transfer station road on a commute route, and a store at a destination.

  Returning to FIG. 1, the mesh group processing unit 111 will be described. The mesh group processing unit 111 executes a mesh grouping process and a group-by-group calculation process. In the mesh grouping process, mesh areas satisfying a predetermined connection relationship are grouped with respect to mesh areas included in the daily life. When the GPS positioning accuracy of the mobile terminal 20 is low, there is a possibility that the mesh area adjacent to the mesh area where the user is actually located is measured as the current position. For this reason, in this embodiment, mesh regions that satisfy a predetermined connection relationship are grouped and processed. The “predetermined connection relationship” can be set as appropriate according to the specifications and design. For example, at least a part of the mesh region is in contact (if the vertices of different mesh regions are in contact, the sides of the mesh region are Including, but not limited to, a case where a predetermined number (for example, 1) of extraordinary sphere meshes are sandwiched between them. The grouped mesh region is referred to as “mesh group”. When there is no mesh area that satisfies the predetermined connection relationship, the mesh area is handled as one mesh group. FIG. 7A is a diagram for explaining grouping of mesh regions. In FIG. 9A, mesh data M11 indicates the daily sphere before grouping, and mesh data M12 indicates the daily sphere after grouping.

  The group calculation process calculates the group stay days, the group stay seconds, and the group positioning times for each mesh group. The number of stays by group is the sum of the stay days of all mesh areas in the same mesh group (however, duplicate days are not double-added), or the maximum stay days of all mesh areas in the same mesh group The number of days of stay. The group stay seconds may be the sum of the stay seconds of all mesh areas in the same mesh group, or the maximum stay seconds among the stay seconds of all mesh areas in the same mesh group. it can. The number of positioning times by group can be the total number of positioning times of all mesh regions in the same mesh group, or the maximum number of positioning times among the positioning times of all mesh regions in the same mesh group. FIG. 7B shows an example in which the staying time per group and the staying time per group are calculated.

  Next, the behavior base likelihood calculation unit 112 identifies a behavior base mesh group from the mesh group on the condition that the mesh group satisfies a predetermined calculation condition, and the mesh region included in the identified behavior base group For each, the probability of being a specific action base of the user (hereinafter referred to as “belief of action base”) is calculated. Specifically, first, it is determined whether or not there is a mesh group satisfying a predetermined calculation condition in the mesh group. If the determination result is negative, in principle, the user's action base is unknown. End the process. Details of the predetermined calculation condition will be described later. On the other hand, if the determination result is correct, the action base mesh group specifying process is executed. The action base mesh group specifying process specifies a mesh group having the maximum number of stay days by group, number of stay seconds by group, or number of positioning times by group from among mesh groups as the action base mesh group. For example, when the action base is “home”, the mesh group having the maximum “stay period by group” is selected as the “home mesh group”. On the other hand, when the action base is “work location”, the mesh group having the maximum “stay time per group” is selected as the “work location mesh group”.

  When the behavior base mesh group is specified, the behavior base likelihood calculation unit 112 determines at least two of the stay days, the stay seconds, and the positioning times for the mesh area included in the behavior base mesh group (action base parameter). Is used according to each priority order to calculate the likelihood of an action base. For example, when the action base is “home”, the home-likeness is calculated using “stay days” (first priority), and when the home cannot be identified by this, “stay seconds” (second priority) The order of home is calculated using the ranking. On the other hand, if the location of action is “work location”, the “likeness of stay” (first priority) is used to calculate the likelihood of work location. The work place likelihood is calculated using the second priority). In the case of “home”, “stay days” has a higher priority than “stay seconds” because the departure point is many at home regardless of the destination, including weekends and holidays. This is because it is assumed to be the maximum. On the other hand, in the case of “work location”, “seconds stayed” has a higher priority than “stay days” because the nearest station or transfer station at home is also used on weekends and holidays. This is because it is assumed that the number of stays may be longer than the ground, while the number of stays is assumed to be larger in the work place. Note that the behavior base-likeness calculation method using the behavior base parameter can be set according to the specification or design, but here, the value of the behavior base parameter is a value indicating the behavior base-likeness. Further, “positioning count” (third priority) may be added for the identification processing of the home and work place.

  The behavior base identifying unit 113 identifies a mesh region that is a user's behavior base based on the behavior base likelihood calculated by the behavior base likelihood calculating unit 112. For example, the mesh region having the highest home-like property can be specified as the user's “home”, and the mesh region having the highest home-like property can be specified as the user's “work location”. FIG. 8A shows an example in which a home mesh group is identified from mesh groups, homeness is calculated for mesh regions in the home mesh group, and a mesh region having the maximum homeness is identified as a home mesh. Is shown. In FIG. 8B, a work area mesh group is identified from the mesh groups, the work area likelihood is calculated for the mesh area in the work area mesh group, and the mesh area having the maximum work place likelihood is the work area mesh. An example identified as

  Next, the structure of each database will be described.

  The positioning information DB 114 stores positioning information transmitted from the mobile terminal 20. FIG. 9A shows an example of the data structure. According to the figure, the positioning information DB 114 stores data such as “positioning time”, “user ID”, “latitude”, and “longitude” in association with each other as a history of positioning information.

  The basic data DB 115 stores the basic data generated by the basic data generation unit 104. FIG. 9B shows an example of the data structure. According to the figure, the basic data DB 115 stores data such as “user ID”, “positioning date”, “mesh number”, and “positioning time” in association with each other. The “mesh number” stores identification information for uniquely identifying a mesh area.

  The daily mesh DB 116 stores the daily mesh data generated by the daily mesh generation means 106. FIG. 9C illustrates an example of the data structure. According to the figure, the daily mesh DB 116 stores data such as “user ID”, “positioning date”, “mesh number”, “staying seconds”, and “number of times of positioning” in association with each other.

  The period-specific mesh DB 117 stores the period-specific mesh data generated by the period-specific mesh generation unit 109. FIG. 10A shows an example of the data structure. According to the figure, data such as “user ID”, “target period”, “mesh number”, “total number of staying seconds”, “total number of positioning times”, “total number of staying days” are stored in the mesh DB 117 by period. Stored in association.

  The daily area DB 118 stores data relating to the daily area setting, and specifically includes daily area conditions used in the daily area setting and data of the set daily area.

  FIG. 10B is a diagram illustrating an example of the daily sphere conditions included in the daily sphere DB 118. Although the contents according to the specification design can be set in the daily sphere conditions, and there is no particular limitation, in the figure, the daily sphere types and conditions are stored in association with each other. Specifically, “Aggregation Period”, “Number of Stay Days”, and “Number of Stays” are set for Daily Area A, and “Target Count Period”, “Positioning Days”, and “Number of Stays” are set for Daily Area B. And “number of times of positioning” are set, and “daily counting period” and “stay days” are set in the daily life zone C. Further, for each item, an arbitrary period and the number of days, time, and number of times as a threshold are set for each type of daily area. As shown in the figure, the conditions for the daily life can be set according to the type of the daily life to be extracted. For example, if you place importance on the number of days you stay and do not consider the number of seconds to stay or the number of positioning, you can specify the daily area that mainly uses the commute route, and if you do not consider the number of days to stay with an emphasis on the number of stays It is possible to specify a daily life area that includes a place where the number of visits is small but stays long (eg, parents' home). In addition, if you set a long period for the target aggregation period, you can prevent the daily area from disappearing when you go on a long trip or a business trip, while if you set a short period for the target aggregation period, It can be easily identified.

  FIG. 10C is a diagram illustrating an example of a user's daily area included in the daily area DB 118. In the figure, for each user ID, “type of daily life” and “mesh number” are stored in association with each other. According to the figure, it can be seen that the mesh number of the set daily sphere differs depending on the type of daily sphere.

  The behavior base DB 119 stores data related to the behavior base identification process. As illustrated in FIG. 11, the behavior base DB 119 includes a calculation condition table that stores calculation conditions for the behavior base and data regarding the behavior base specified for each user. It includes a coefficient table for storing action base data to be stored and coefficients for each positioning level used when calculating the likelihood of the action base.

  In the calculation condition table, conditions according to the characteristics of the action base can be set. In FIG. 11A, for example, in the case of home, “the number of stay days by group is not less than a predetermined number of days (n days)”. In the case of a work place, “the number of stays by group is equal to or greater than the predetermined number of days (x seconds)”, and , The number of stay days by group is equal to or greater than a predetermined number of days (y days). n is the minimum number of days stayed (eg, n is 20 days), m is the minimum length of stay (eg, m is 432000 seconds = 15 days × 8 hours × 60 minutes × 60 seconds), x is the minimum working days (eg, x is 234000 seconds = 13 days × 5 hours × 60 minutes × 60 seconds), and y is the minimum working time (for example, y is 13 days).

  As shown in FIG. 11B, the action base data includes data such as “user ID”, “type of action base”, “mesh number” of the action base, and “date” specifying the action base. , Stored as a history of action bases.

  In the coefficient table, as shown in FIG. 11C, “positioning level” and “coefficient” are stored in association with each other. The number of positioning levels and the coefficient values can be set according to specifications and design. In addition, the calculation process of the likelihood of an action base using a positioning level will be described in other embodiments.

[2. Configuration of mobile terminal]
Returning to FIG. 1, the mobile terminal 20 will be described. As the mobile terminal 20, for example, a conventional terminal device having a function of uploading positioning information obtained by measuring the current position at predetermined time intervals using GPS such as a mobile phone, a PDA, and a personal computer can be applied. Although one mobile terminal 20 is illustrated in FIG. 1, a plurality of mobile terminals 20 can be connected to the behavior base identifying system 10 according to the usage mode.

  As shown in FIG. 12, the mobile terminal 20 includes various function realizing means such as a main control means 201, a communication means 202, a display means 203, an operation means 204, a storage means 205, a current position positioning means 206, and a positioning information transmission means 207. Prepare for the main.

  The main control unit 201 includes a CPU (not shown) and a processor including a memory such as a ROM and a RAM, and controls the operation of each unit of the portable terminal 20 by the CPU executing a predetermined program stored in the ROM. . The communication unit 202 is an interface for transmitting and receiving various types of information to and from the action base specifying system 10 via the network N. The display unit 203 is a display that displays information such as characters and images, and the operation unit 204 is a button or a touch panel that receives an operation instruction from the user. The storage unit 205 is a memory as a storage device that stores various programs and data.

  The current position positioning means 206 includes, for example, a GPS receiver, and measures the current position (latitude / longitude) of the mobile terminal 20 by receiving and processing GPS satellite signals at predetermined reception intervals.

  The positioning information transmitting unit 207 transmits positioning information including the positioning point and positioning time determined by the current position positioning unit 206 and the user ID of the user who owns the mobile terminal 20 to the behavior base identifying system 10. The positioning information can be appropriately set according to the specification / design. However, as described above, in this embodiment, when the user is moving, the positioning information is transmitted according to a predetermined transmission interval. It is configured as follows. Whether or not the user is moving can be determined, for example, by applying a conventional technique using an acceleration sensor (not shown).

  The network N is a communication line for transmitting and receiving information between the behavior base specifying system 10 and the mobile terminal 20. For example, it may be any of the Internet, a LAN, a dedicated line, a packet communication network, a telephone line, a corporate network, other communication lines, combinations thereof, and the like, regardless of whether they are wired or wireless.

[3. Action area identification process]
Next, the outline | summary of operation | movement of the action base identification system 10 comprised as mentioned above is demonstrated, referring the flowchart of FIGS. In addition, each processing step to be described later can be executed in any order or in parallel as long as there is no contradiction in processing contents, and other steps may be added between the processing steps. . Further, a step described as one step for convenience can be executed by being divided into a plurality of steps, while a step described as being divided into a plurality of steps for convenience can be grasped as one step.

[3-1. Daily zone setting process]
First, the daily zone setting process will be described. FIG. 13 is a flowchart showing an example of the overall flow of the daily area setting process. When receiving the positioning information (positioning point, positioning time, user ID) transmitted from the mobile terminal 20 at a predetermined transmission interval, the behavior base specifying system 10 stores the received positioning information in the positioning information DB 114 (S100). Thereby, the history of the positioning information according to the movement of the user is stored in the positioning information DB 114 (see FIG. 3A).

  The behavior base identifying system 10 rearranges the positioning information in the positioning information DB 114 in time series according to the date for each user, and generates basic data in which the positioning point and the positioning time are associated with the corresponding mesh region for each positioning date for each user (S200). (See FIG. 6). The generated basic data is stored in the basic data DB 115 (see FIG. 3B).

  Based on the generated basic data, the behavior base identifying system 10 generates daily mesh data representing the number of times of positioning in each mesh region and the number of seconds the user stays for each predetermined unit period (eg, 1 day) for each user. (S300). The generated daily mesh data is stored in the daily mesh DB 116 (see FIGS. 4A and 7). Details of the daily mesh data generation process will be described later.

  Based on the generated daily mesh data, the behavior base identifying system 10 totals the number of positionings, the number of seconds of stay, and the number of days of stay in each mesh area in a predetermined counting period, Period-specific mesh data in which the number of seconds and the number of stay days are associated is generated (S400). The generated period-specific mesh data is stored in the period-specific mesh DB 117 (see FIG. 10).

  The behavior base identifying system 10 is a mesh region in which any one of the number of positioning times, the number of staying seconds, and the number of staying times included in the period-specific mesh data, or an arbitrary combination of two or more thereof satisfies a predetermined daily sphere condition. Is extracted from the user's daily sphere (S500). The generated daily life data is stored in the daily life DB 118 (see FIGS. 4B and 4C and FIG. 11).

[3-1-1. Daily mesh data generation process]
Next, the flow of daily mesh data generation processing will be described with reference to FIG. First, the action base specifying system 10 sets, based on basic data, a set of a first positioning point of a positioning point before movement and a second positioning point of a positioning point after movement from a positioning point on a mesh area, for example, in time series. If the group can be identified (S301: YES), the process proceeds to step 302 to apply the first distribution condition or the second distribution condition. If the group cannot be identified (S301: NO) ), Go to step 309 to apply the third distribution condition.

  When the action base identification system 10 proceeds to step 302, the action base identification system 10 calculates the distance between the first positioning point and the second positioning point, the first positioning time of the first positioning point, and the second positioning point second. The time between positioning points between the positioning times is calculated (S302).

  As an example of the allocation condition, the behavior base identifying system 10 determines whether the calculated distance between positioning points is less than a predetermined reference distance and whether the calculated time between positioning points is less than a predetermined reference time (S303). If the determination result is right, the process proceeds to step 304 to apply the first distribution condition. If the determination result is negative, the process proceeds to step 307 to apply the second distribution condition.

  When the action base identification system 10 proceeds to step 304, the action base identification system 10 sets the transmission time interval of the positioning information in the mobile terminal to the travel time (S304). Then, the travel time subtracted time is calculated by subtracting the travel time from the positioning point time, and the calculated travel time subtracted time is allocated to the first mesh region to which the first positioning point belongs as the allocated staying seconds. (S305). Next, the moving time is distributed approximately evenly as the allocated staying seconds to one or a plurality of mesh regions located on the connection between the first positioning point and the second positioning point (S306). As a result, the allocated stay seconds are appropriately distributed to the first mesh region, the second mesh region, and the complementary mesh region (see FIG. 8A).

  On the other hand, when the action base identification system 10 proceeds to step 307, it calculates the travel time based on the distance between the positioning points and the predetermined speed information (S307). Then, the travel time subtracted time is calculated by subtracting the travel time from the positioning point time, and the calculated travel time subtracted time is allocated to the first mesh region to which the first positioning point belongs as the allocated staying seconds. (S308). In addition, the allocation stay seconds are not distributed to the second mesh area to which the second positioning point belongs or other mesh areas. Thereby, the allocation staying seconds are appropriately allocated to the first mesh area (see FIG. 8B).

  In addition, since the judgment result is NO in step S301, the action base identification system 10 is the last positioning in which the first positioning point in which the target positioning point is first positioned in the corresponding day is finally determined when the process proceeds to step 309. If it is the first measurement point, the time from the reference time 0:00 to the positioning time of the target positioning point is distributed to the target mesh area within a predetermined maximum time (S310). ). On the other hand, when the target positioning point is the last positioning point, the time from the positioning time of the target positioning point to 24 o'clock as the reference time is distributed to the target mesh area within a predetermined maximum time (S311). . Thereby, the allocation staying seconds are appropriately allocated to the mesh areas of the first positioning point and the last positioning point (see FIG. 9).

  The behavior base specifying system 10 calculates the staying seconds in the target mesh area based on the allocated staying seconds allocated to the mesh area, and the number of times the positioning point is positioned in the mesh area is calculated. (S312). For the complementary mesh region, the staying seconds are added, but the positioning point is not added.

  Next, the behavior base identifying system 10 determines whether or not there is a positioning point to be processed next. If there is a positioning point to be processed next (S313: YES), the process returns to step S301. Repeat the above process. On the other hand, if there is no positioning point to be processed next, the process is terminated (S313: NO).

[3-1-2. Periodic mesh data generation process]
Next, the flow of the period-specific mesh data generation process will be described with reference to FIG. 15A (see FIG. 10). The action base identification system 10 determines whether or not the period-specific mesh data for the target aggregation period exists in the period-specific mesh DB 117, and if it has already been generated (S401: YES), the process proceeds to step S402 and is still generated. If not (S401: NO), the process proceeds to step S403.

  When the action base identification system 10 proceeds to step S402, the existing period-specific mesh data is read from the period-specific mesh DB 117, the daily target mesh data is read from the daily mesh DB 116, and the daily target-subtraction date is subtracted. For each mesh area, the number of staying seconds, the number of positioning times, and the number of staying days of the addition target day are added, while the number of staying seconds, the number of positioning times, and the number of staying days of the subtraction target day are subtracted (S402).

  When the action base specifying system 10 proceeds to step S403, the daily base mesh data for the target totaling period is read from the daily mesh DB 116, and the staying seconds, positioning times, and staying days for the target totaling period are totaled ( S403).

  The action base identification system 10 stores the generated period-specific mesh data in the period-specific mesh DB 117 (S404).

[3-1-3. Daily zone setting process]
Next, the flow of the daily area setting process will be described with reference to FIG. 15B (see FIG. 11). The action base identification system 10 refers to the daily sphere conditions of the target daily sphere type from the daily sphere DB 118 (S501), and reads the corresponding period-specific mesh data from the period-specific mesh DB 117. A matching mesh region is extracted and set as a daily zone (S502). The action base specifying system 10 associates the set type of the daily sphere with the mesh number and stores it in the daily sphere DB 118 (S503).

[3-2. Action base identification process]
Next, action base specifying processing for specifying the user's action base using the daily zone set as described above will be described. Here, as an example, a case where “home” is specified as the first action base and “work location” is specified as the second action base will be described as an example. FIG. 16 shows the overall flow of the action base specifying process.

  In other words, the behavior base identifying system 10 generates mesh groups by grouping mesh areas satisfying a predetermined connection relationship from mesh areas included in the daily areas (hereinafter referred to as “daily area meshes”) ( S1000: See FIG. 7A). Next, for each mesh region in each mesh group, the number of stays by group, the number of stays by group, and the number of positioning times by group are calculated (S1010: see FIG. 7B).

  Then, as a first action base, a homeness calculation process is executed for the corresponding mesh area, and a home mesh area specifying process is executed (S1020). Next, as the second action base, the work area likelihood calculation process is executed for the corresponding mesh area, and the work place mesh area specifying process is executed (S1021). Details of these processes will be described later.

[3-2-1. Home mesh area identification process]
FIG. 17 shows the overall flow of the home mesh area specifying process. The behavior base likelihood calculation unit 112 refers to the home calculation condition from the behavior base DB 119 (S701), and the home calculation condition (for example, stay days by group ≧ minimum stay days n by group) among the generated mesh groups and by group. It is determined whether or not there is a mesh group satisfying the number of stay seconds ≧ minimum stay time seconds m) (S702). If the determination result is negative (S702: NO), it is determined that the home mesh area is unknown for the user (S703).

  On the other hand, if the determination result is right (S702: YES), the mesh group having the largest “stay time per group” is identified as the home mesh group from the mesh groups that match the home calculation conditions (S704). When a plurality of home mesh groups having the maximum stay days are specified (S705: YES), the mesh group having the largest “stay time by group” is selected as the home mesh group (S706). Further, when a plurality of home mesh groups having the maximum staying seconds are specified (S707: YES), the mesh group having the largest “number of positioning by group” is selected as the home mesh group (S708).

  When the home mesh group is specified, the behavior base likelihood calculation unit 112 calculates a value indicating the homeness for each mesh in the home mesh group using the behavior base parameter. Here, “stay days” (first priority), “stay seconds” (second priority), and “number of times of positioning” (third priority) are set in the action base parameters together with the priority. Therefore, first, a value indicating homeness is calculated using the “stay days” of each mesh region (S709).

  When the value indicating the homeness is calculated for all the mesh regions in the home mesh group, the behavior base identifying unit 113 determines the mesh region having the largest value indicating the homeness as the home mesh region (S710). If a plurality of home mesh areas having the maximum stay days are specified (S711: YES), a value indicating homeness is calculated based on the second priority "stay seconds", and the value is the largest. The mesh area is selected as the home mesh area (S712). Further, when a plurality of home mesh areas having the maximum staying seconds are specified (S713: YES), a value indicating homeness is calculated based on the “number of times of positioning” that is the third priority, and the value is the largest. The mesh area is selected as the home mesh area (S714).

  When the home mesh area of the user is determined, the behavior base identifying unit 113 executes a predetermined home confirmation process (S715), and then associates the user with the mesh number of the home mesh area and stores it in the behavior base DB 119. (S716). Details of the home confirmation process will be described later. Through the above process, the user's home mesh area is identified from the daily life.

[3-2-2. Work area mesh area identification process]
FIG. 18 shows the overall flow of the work location mesh area specifying process. The behavior base likelihood calculation unit 112 excludes the home mesh group from the target mesh group, and refers to the work location calculation condition from the behavior base DB 119 (S801). Then, among the target mesh groups after excluding the home mesh group, the mesh group that satisfies the work location calculation condition (for example, the stay days by group ≧ minimum work days x and the stay seconds by group ≧ minimum work seconds y) It is determined whether or not there is (S802).

  If the determination result is right (S802: YES), a mesh group having the largest “stay time per group” is identified as a work place mesh group from mesh groups that match the work place calculation condition (S804). If a plurality of work place mesh groups having the maximum stay seconds are specified (S805: YES), the mesh group having the largest “stay days by group” is selected as the work place mesh group (S806). Furthermore, when a plurality of work location mesh groups having the maximum stay days are specified (S807: YES), the mesh group having the largest “number of positioning by group” is selected as the work location mesh group (S808).

  When the work location mesh group is specified, a value indicating the likelihood of work location is calculated for each mesh in the work location mesh group using the behavior base parameter. Here, “staying seconds” (first priority), “staying days” (second priority), and “number of times of positioning” (third priority) are set together with the priority in the action base parameter. Therefore, first, a value indicating the likelihood of work is calculated using “stay seconds” of each mesh region (S809).

  When the value indicating the work place likelihood is calculated for all mesh areas in the work place mesh group, the action base specifying unit 113 determines the mesh area having the largest value indicating the work place likelihood as the work place mesh area ( S810). When a plurality of work place mesh areas having the maximum stay seconds are specified (S811: YES), a value indicating work place likelihood is calculated using the second priority “stay days” as an action base parameter. The mesh area having the maximum value is selected as the work area mesh area (S812). Further, when a plurality of work place mesh areas having the maximum stay days are specified (S813: YES), a value indicating the work place likelihood is calculated using the third priority "positioning count" as an action base parameter, and the value The mesh area having the largest is selected as the work place mesh area (S814).

  When the user's work place mesh area is determined, the action base specifying unit 113 associates the user with the ID of the work place mesh area, and stores it in the action base DB 119 (S812). If it is determined in step S802 that there is no mesh group that satisfies the work location calculation condition (S802: NO), the behavior base likelihood calculation means 112 executes a predetermined work location confirmation process (S813), and the work location confirmation If it is determined that the work location is unknown even after the processing (S814: NO), it is determined that the work location mesh area is unknown for the user (S815). On the other hand, if it is determined by the work place confirmation process that there is a work place (S814: YES), the process proceeds to step 812, and the user and the mesh number of the work place mesh area are associated and stored in the action base DB 119 ( S812). Through the above processing, the work location mesh area of the user is specified from the daily life.

[3-2-3. Home confirmation process]
Next, the home confirmation process will be described. For example, when the user has just moved, the stay days may be distributed over a mesh area including the home before moving and the home after moving (see FIG. 19A). As a result, the number of stays in the mesh area including the work place may be larger than the stay days in the mesh area including the home after moving, and the work place may be determined as the home and the home after moving may be determined as the work place. . Therefore, in this embodiment, in order to identify an appropriate user's action base in consideration of such a change in the action base, a confirmation process for reconfirming the result of the home determination process described above is provided. FIG. 20 shows the overall flow of the home confirmation process.

  When the home mesh area is specified (FIG. 17: S711), the behavior base specifying unit 113 executes the home confirmation process. That is, referring to the mesh number of the work place mesh area (hereinafter referred to as “reference work place mesh area”), which is a work place mesh area specified in the past and matches a predetermined reference condition, from the action base DB 119, It is determined whether or not the mesh number of the home mesh area specified this time matches the mesh number of the reference work area mesh area (S901). As the predetermined reference condition, for example, contents such as “work location mesh region specified last time” or “work location mesh region specified continuously more than a predetermined number of times” can be set.

  If the determination result is NO (= no match), the behavior base specifying unit 113 ends the process and proceeds to step S713 in FIG. 17 (S902: NO). On the other hand, when the determination result is “good” (= match), the work location mesh area is determined as the current work place mesh area, assuming that the work place is erroneously determined as the home (S903). Then, the process moves to step S701 in FIG. 17 so that the mesh group including the reference work area mesh region is excluded from the target mesh group, and the home mesh region specifying process is re-executed for the target mesh group after the exclusion (S904). ).

  According to the above, even if the work location is mistakenly determined as a home due to moving, etc., the error is detected and the home mesh area specifying process is re-executed. Will be able to decide. Note that the home confirmation process is not limited to the case where the workplace is erroneously determined as home, but can be applied when there is a possibility that one action base may be erroneously determined as another action base.

[3-2-4. Work location confirmation process]
Next, work location confirmation processing will be described. For example, when the user's work location is located near his / her home, the work location mesh area may be included in the home mesh group. Since it is excluded, the work area mesh region may not be specified (see FIG. 19B). Therefore, in the present embodiment, in consideration of the positional relationship between the home and the work place, a confirmation process for reconfirming whether or not the user's work place exists is provided.

  When it is determined that the work location mesh area is unknown because there is no mesh group that matches the work location calculation condition (S812 in FIG. 18: NO), the behavior base likelihood calculation unit 112 executes the work location confirmation process. FIG. 21 shows the overall flow of work location confirmation processing.

  First, the behavior base likelihood calculation unit 112 sets a mesh region other than the home mesh region in the home mesh group as a target mesh region (S1001). Then, it is determined whether or not the target mesh region satisfies the predetermined stay days, stay seconds, and positioning times (first extraction criterion) (S1002).

  If the determination result is positive (S1002: YES), it is determined whether or not the target mesh region that satisfies the first extraction criterion is a predetermined distance away from the home mesh region (second extraction criterion) (S1003). If the determination result is positive (S1003: YES), for the target mesh region that satisfies the second extraction criterion, the difference in the number of stays between the target mesh region and the home mesh region, the difference in the number of staying seconds, the difference in the number of positioning times Is a predetermined threshold (third extraction criterion) or not (S1004).

  If the determination result is correct (S1004: YES), whether or not a specific facility (eg, company, factory, school, etc.) is included for the target mesh region that satisfies the third extraction criterion (fourth extraction criterion). ) Is determined (S1005). Then, when the determination result is “good” (S1005: YES), for the target mesh region satisfying the fourth extraction criterion, the work place likelihood is calculated using the behavior base parameter (here, the number of staying seconds) (S1006). .

  When the value indicating the work place likelihood is calculated for the mesh area that matches the first extraction criterion to the fourth extraction reference, the action base specifying unit 113 determines the mesh area having the largest work place likelihood as the work place mesh region. (S1007).

  On the other hand, when the first extraction criterion to the fourth extraction criterion are not satisfied (S1002: NO, S1003: NO, S1004: NO, S1005: NO), the action base specifying unit 113 determines that the work location mesh region is unknown. .

  According to the above, for example, even when the work place and the home are included in the same mesh group, the work place can be appropriately specified. The extraction criteria for the target mesh can be set according to the specifications and design, and the extraction criteria can be deleted, added, or changed for the first to fourth extraction criteria described above. The work location confirmation process is not limited to the relationship between the work location and the home, but can be applied when one action base may be included in a mesh group of another action base.

  As described above, according to the present embodiment, since the user can automatically set the action base of the user simply by carrying the portable terminal with the GPS function, various information registration work for setting the action base can be performed. It becomes unnecessary, and the convenience for the user can be improved. In addition, the behavior base is calculated by using at least two parameters of the number of positioning times, the number of stays, and the number of days of stay according to the priority order according to the content of the behavior base. It becomes possible to increase the accuracy of the.

[Examples of use of action bases]
An example of using the action base set as described above will be described. The usage method of the action base can be determined according to the purpose, etc., and there is no particular limitation on the content thereof. The content of the information to be provided can be determined after considering the relationship between the facility and the user's action base. For example, when the target facility is located in the vicinity of the user's home / work location mesh area or in the home / work location mesh group, the sale information, lunch information, real estate information, etc. Can be provided.

  In addition, the mesh area in the mesh group excluding the home mesh group and work place mesh group from the daily life area is identified as the “frequently meshed area” as the third action base, and is located in the identified “frequently meshed area” It is also possible to discriminate the user's hobbies (eg, golf, skiing) from information (eg, golf courses, ski resorts) regarding the facility to be performed, and provide information on the discriminated hobbies.

  Furthermore, it is possible to determine the occupation and attributes of the user by using the specified action base mesh group and action base mesh area. For example, a user for whom a work location mesh group is specified can be determined as a company employee or a student, and a user for whom a work location mesh group has not been specified can be determined as a housewife or a self-employed person. In addition, when the work location mesh area includes a university, the user is determined to be a college student, and when the work location mesh group is located in an office district, the user can be determined to be a company employee.

[Other Embodiments]
The present invention is not limited to the above-described embodiment, and can be implemented in various other forms without departing from the gist of the present invention. The above-described embodiment is merely an example in all respects, and is not construed as limiting.

(1) For example, the behavior base likelihood calculation unit 112 may calculate a value indicating the behavior base likelihood by multiplying the value of the behavior base parameter by a coefficient corresponding to the positioning level. The positioning level indicates an error range of the positioning result. For example, the positioning level 3 is “horizontal error <50 m”, the positioning level 2 is “50 m ≦ horizontal error <300 m”, and the positioning level 1 is “300 m ≦ horizontal error”. Is shown. As described above, since the error becomes larger as the positioning level is lower, there is a possibility that there is actually no user in the mesh area indicated by the positioning information. Therefore, the coefficient is set so that the weight becomes smaller as the positioning level is lower (see FIG. 11C). Then, for each mesh region in the action base mesh group, the formula = {(positioning level 1 positioning times × positioning level 1 coefficient × action base parameter) + (positioning level 2 positioning times × positioning level 2 coefficient × action Base parameter) +... + (Positioning frequency of positioning level n × coefficient of positioning level n × action base parameter)}, and the result of the calculation is a value indicating the likelihood of the action base of the mesh region. According to this, since the value indicating the likelihood of the behavior base can be calculated in consideration of the positioning level, it is possible to further improve the identification accuracy of the behavior base mesh region. The coefficient setting includes a case where 1 is set for the coefficients of all the positioning levels, and in this case, it is substantially the same as not considering the positioning level.
(2) The behavior base likelihood calculation means 112 may obtain a sum of weights by assigning a larger weight to the value of each behavior base parameter that has a higher priority, and use it as a value indicating the likelihood of the behavior base.

  (3) The behavior base attributes can be determined using the behavior base parameters “stay seconds” and “positioning count”. When the mobile terminal 20 is configured to transmit positioning information only when the user moves, the number of positioning increases if the user moves, while the number of positioning decreases when the user does not move much. Accordingly, when a certain mesh area has a small number of staying seconds but a large number of positioning times, it can be assumed that the mesh area is an active action base (for example, a shopping mall). On the other hand, when the number of staying seconds is large but the number of positioning times is small, the mesh area can be assumed to be an inactive action base (for example, a movie or a concert).

  (4) For example, when it is determined that the positioning accuracy of the current position in the mobile terminal 20 is low, a mesh area (hereinafter referred to as “movement path mesh”) located on the movement path from the first positioning point to the second positioning point. One or a plurality of mesh regions located in the vicinity of the “region” may be set as the peripheral mesh region, and the time between positioning points may be allocated to the peripheral mesh region. The surrounding mesh area can be set as appropriate according to the design and specifications. For example, when the positioning accuracy of the first positioning point and / or the second positioning point is low, the first mesh area and / or Alternatively, the mesh area around the second mesh area can be set as the first peripheral mesh area and / or the second peripheral mesh area. For example, when the positioning accuracy of both the first positioning point and the second positioning point is low, a mesh area around the complementary mesh area on the movement path may be set as the complementary peripheral mesh area. When the surrounding mesh area is set, the distribution complementing means 108 is calculated according to the distributing method described in the above embodiment, for example, and the movement route mesh area corresponding to the surrounding mesh area (eg, the first mesh surrounding area). The allocation stay seconds for (example: first mesh area) can be distributed to both the movement path mesh area and the peripheral mesh area (example: first mesh area and first mesh peripheral area). According to this configuration, when a positioning point with low accuracy is received, the number of staying seconds is also distributed around the positioning point. As a result, it is possible to improve the accuracy of the daily life zone.

  DESCRIPTION OF SYMBOLS 10 ... Action base identification system, 20 ... Mobile terminal, 101 ... Main control means, 102 ... Communication means, 103 ... Positioning information storage means, 104 ... Basic data generation means, 105 ... mesh data generation means, 106 ... daily mesh generation means, 107 ... positioning count counting means, 108 ... distribution complementing means, 109 ... period-specific mesh generation means, 110 ... daily area setting Means 111... Mesh group processing means 112... Behavior base-likeness calculation means 113... Action base identification means 114... Positioning information DB 115 115 Basic data DB 116. Daily mesh DB, 117 ... Period mesh DB, 118 ... Daily service area DB, 119 ... Behavior base DB, 201 ... Main control means, 202 ... Communication means, 203. And display means, 204 ... operating unit, 205 ... storage unit, 206 ... current position measuring means, 207 ... positioning information transmission means, N ... Network

Claims (11)

Storage means for associating the positioning point obtained by measuring the current position of the mobile terminal transmitted from the mobile terminal with the positioning time of the positioning point in a predetermined database;
Basic data in which the positioning point and the positioning time stored in the predetermined database are associated with a mesh region to which the positioning point belongs among a plurality of mesh regions obtained by dividing a map into a mesh. Basic data generation means for generating
Based on the generated basic data, among the number of positioning days in which the positioning point is positioned in each mesh area within a predetermined period, the number of times the positioning point is positioned, and the staying time during which the user of the mobile terminal stayed Are calculated as behavior base parameters, and mesh data in which each mesh region is associated with the calculated two behavior base parameter values are generated for each positioning date, and the mesh data for each positioning date is aggregated. Mesh data generating means for generating mesh data for the period by
A behavior base likelihood calculating means for calculating the behavior base likelihood of the mesh region by using the values of the two behavior base parameters for the generated period according to respective priorities;
An action base specifying means for specifying the mesh area having the largest calculated action base likelihood as the action base mesh area including the user's action base from the plurality of mesh areas;
An action base identification system characterized by comprising: A daily area setting means for setting, as the daily area of the user, a mesh area in which a value of two action base parameters for the generated period is equal to or greater than a predetermined daily area threshold from the plurality of mesh areas;
Group processing means for grouping mesh regions satisfying a predetermined connection relationship for mesh regions included in the set daily sphere, and calculating group-specific action base parameters for each grouped mesh group ,
The behavior base likelihood calculation means includes:
Identifying an action base mesh group from the mesh group based on the calculated group-specific action base parameters, calculating the likelihood of an action base for each mesh region in the specified action base mesh group,
The action base specifying means is:
2. The mesh area having the highest likelihood of the calculated action base is specified as the user's action base mesh area from among the mesh areas in the specified action base mesh group. Action base identification system. When the user's action base is the first action base that is the home of the user's life, the two action base parameters are set with the number of days of positioning having the first priority and the staying time having the second priority. And
The behavior base likelihood calculation means includes:
From among the mesh groups, a mesh group that satisfies the predetermined first action base calculation condition and has the maximum number of positioning days by group as the first action base mesh group is specified as the first action base mesh group. In addition, for each mesh region in the identified first action base mesh group, a value indicating the first action base characteristic is calculated by using the positioning days and the staying time according to each priority. The action base identification system according to claim 2. When the user's action base is the second action base where the user stays continuously, the two action base parameters are set with the stay time having the first priority and the positioning days having the second priority. ,
The behavior base likelihood calculation means includes:
Among the mesh groups, a mesh group that satisfies the predetermined second action base calculation condition and has the maximum stay time by group as the second action base mesh group is specified as the positioning days by group and the stay time by group. In addition, for each mesh region in the identified second action base mesh group, a value indicating the second action base characteristic is calculated by using the stay time and the positioning days in accordance with each priority. The action base identification system according to claim 2. If multiple locations are identified,
The action base specifying means is:
If the action base mesh area of one specified action base matches the action base mesh area of another action base identified in the past, the mesh group including the action base mesh area of the other action base is excluded. The behavior base identification system according to claim 2, wherein the behavior base likelihood calculation unit is controlled to calculate again the behavior base likelihood of the one behavior base for the mesh group. If multiple locations are identified,
The behavior base likelihood determination means includes:
When a mesh group of one action base can be specified and a mesh group of another action base cannot be specified, a mesh other than the action base mesh area of the one action base included in the mesh group of the one action base. The behavior base identifying system according to claim 2, wherein the behavior base likelihood of the other behavior base is calculated for a region. The mesh data generation means includes
For the second positioning point arranged next to the first positioning point and the first positioning point on the time axis, the positioning point from the first positioning time of the first positioning point to the second positioning time of the second positioning point The positioning time is calculated according to the allocation condition determined based on the time between the positioning points and the distance between the positioning points, and calculates the distance between the positioning points between the second positioning point and the first positioning point. Distributing means for allocating the time between points as stay time in a mesh region on the connection connecting the first positioning point and the second positioning point;
The behavior base identifying system according to claim 1, wherein the stay time in the mesh area is calculated based on the allocated allocation time. The distribution means includes
When the time between the positioning points is less than a predetermined time and / or the distance between the positioning points is less than the predetermined distance, the moving time from the first positioning point to the second positioning point is subtracted from the time between the positioning points. 8. The movement time subtracted time obtained in the above is distributed to the first mesh area to which the first positioning point belongs, and the movement time is substantially evenly distributed to the mesh area on the connection. Action base identification system. The distribution means includes
When the time between the positioning points is less than a predetermined time and / or the distance between the positioning points is not less than the predetermined distance, the moving time from the first positioning point to the second positioning point is subtracted from the time between the positioning points. 9. The action base according to claim 8, wherein the movement time subtracted time obtained in the step is distributed to the first mesh area to which the first positioning point belongs, and the movement time is not distributed to the mesh area on the connection. Specific system. The distribution means includes
When the first positioning time of the first positioning point and the second positioning time of the second positioning point extend over a predetermined reference time, the first mesh is set to the maximum time from the first positioning time to the predetermined reference time. 10. The behavior base identification system according to claim 7, wherein the behavior base identification system according to claim 7 is allocated to an area and allocates the time from the second positioning time to the predetermined reference time as a maximum to the second mesh area. Storage means for associating the positioning point obtained by measuring the current position of the mobile terminal transmitted from the mobile terminal with the positioning time of the positioning point in a predetermined database;
Basic data in which the positioning point and the positioning time stored in the predetermined database are associated with a mesh region to which the positioning point belongs among a plurality of mesh regions obtained by dividing a map into a mesh. Basic data generation means for generating
Based on the generated basic data, among the number of positioning days in which the positioning point is positioned in each mesh area within a predetermined period, the number of times the positioning point is positioned, and the staying time during which the user of the mobile terminal stayed Are calculated as behavior base parameters, and mesh data in which each mesh region is associated with the calculated two behavior base parameter values are generated for each positioning date, and the mesh data for each positioning date is aggregated. Mesh data generating means for generating mesh data for the period by
Among the plurality of mesh areas, a mesh area having the largest value of at least one action base parameter in accordance with the priority order among two action base parameters for the generated period is included in the action base including the user action base. Action base identification means to identify as a mesh area,
An action base identification system characterized by comprising: