WO2011102541A1

WO2011102541A1 - Behavior feature extraction device, behavior feature extraction system, behavior feature extraction method, behavior feature extraction program

Info

Publication number: WO2011102541A1
Application number: PCT/JP2011/054063
Authority: WO
Inventors: 岳夫大野
Original assignee: 日本電気株式会社
Priority date: 2010-02-19
Filing date: 2011-02-17
Publication date: 2011-08-25
Also published as: JPWO2011102541A1

Abstract

A home location and a workplace location, which characterize a user's behavior, are extracted from a location information history. A behavior feature extraction device (300) is provided with a behavior type storage unit (202) and a behavior feature extraction unit (302). The behavior type storage unit (202) stores lodging location information containing the lodging locations where a user has lodged, the start dates of lodging at said lodging locations, and the end dates of lodging at said lodging locations. The behavior feature extraction unit (302), on the basis of this lodging location information in the behavior type storage unit (202), calculates the number of days lodged at each of the lodging locations, extracts from said lodging locations the lodging location with the highest number of lodging days as the home lodging location, and extracts from said lodging locations the lodging location with the second highest number of lodging days as the workplace lodging location.

Description

Behavior feature extraction device, behavior feature extraction system, behavior feature extraction method, and behavior feature extraction program

The present invention relates to a behavior feature extraction device, a behavior feature extraction system, a behavior feature extraction method, and a behavior feature extraction program that extract feature information of user behavior.

2. Description of the Related Art In recent years, it has become possible to periodically acquire user location information from a terminal equipped with a GPS (Global Positioning System). As a service using such position information, for example, a service for displaying a map around the user or guiding a route from the current location to the destination is provided. These services are provided based on the current location information of the user. As a further developed service, it has been studied to extract a user behavior type based on a history of user location information and to provide a service based on the extracted behavior type.
For example, Patent Literature 1 discloses a situation estimation device that defines a situation transition model of a behavioral state (sleeping, working, going out, etc.) and estimating a user's current situation from position information and time information. Yes. In the situation estimation apparatus of Patent Document 1, the user needs to input user behavior feature information such as a home position or a work position. For this reason, the user has to register the location information of the nearest station, home, etc., and when the location information is changed, it is necessary to update the location information each time. In addition, operators that provide services to users using the estimated situations provide services on the assumption that the behavioral feature information that has been input may not be updated and may be inaccurate. There was a need to do.
For example, Patent Literature 2, Patent Literature 3, and Patent Literature 4 disclose a method of automatically extracting such behavior feature information of the user instead of the user inputting it.
The behavior history analysis apparatus of Patent Literature 2 estimates a user's private-related point and business-related point based on a staying time zone, a noise level, and an illuminance level at the staying place of the user.
The characteristic extraction device of Patent Document 3 analyzes the visit history data of a moving body for each place according to the characteristic extraction rule, and extracts a frequently visited place or a home place.
The destination prediction apparatus of Patent Document 4 accumulates the movement history of a moving body and determines the home based on the frequency of arrival and the staying time.
Further, as related technology, Patent Document 5 discloses a navigation device that accumulates arrival and departure histories of vehicles and detects changes in home positions based on changes in the accumulation frequency of arrival and departure histories for home positions. .
As another related technique, Patent Document 6 discloses a destination prediction device that acquires boarding / alighting data of a vehicle and detects a permanent place based on variations in boarding time.

JP 2009-181476 A JP 2009-043057 A JP 2000-155757 A JP 2009-036594 A JP 2009-276112 A JP 2006-308382 A

The behavior history analysis device of Patent Document 2 described above classifies whether the staying place is a private related point or a business related point based on the illuminance level and the noise level, and further, the staying time exists in a specific time zone. Depending on whether or not the classification result is corrected. As described above, in the behavior history analysis apparatus of Patent Document 2, in order to classify whether the staying place is a private-related point or a business-related point, information on illuminance and noise other than position information, and a user's working hours There was a problem of needing information about.
In Patent Document 3 and Patent Document 4 described above, a method for extracting a home position based on a history of position information of a moving object is disclosed, but a method for extracting a work position is not disclosed. .
One object of the present invention is to extract a home feature and a work location that characterize a user's behavior from a location information history without using environmental information such as illuminance and noise and information about the user's working hours. To provide an apparatus, a behavior feature extraction system, a behavior feature extraction method, and a behavior feature extraction program.

The behavior feature extraction apparatus according to the aspect of the present invention stores behavior point information that includes each residence point of a plurality of residences of a user, a residence start date and time at the residence point, and a residence end date and time at the residence point. And a staying day for each of the plurality of staying points based on the staying point information of the means and the behavior type storage means, and among the plurality of staying points, the staying point having the most staying days is set at home. It is extracted as a stay point, and is provided with behavior feature extraction means for extracting the stay point having the second most stay days among the plurality of stay points as a workplace stay point.
The behavior feature extraction system according to an aspect of the present invention includes a terminal having a position information acquisition unit that acquires a positioning point indicating a position of a user together with a positioning date and time, and outputs position information including the positioning point and the positioning date and time, Based on the position information, the behavior of extracting the residence point information including each residence point of the plurality of residences of the user, the residence start date and time at the residence point, and the residence end date and time at the residence point, and storing them in the behavior type storage unit A behavior type extraction device having a type extraction means; and the stay point information is acquired from the behavior type extraction device, and based on the stay point information, a staying day is calculated for each of the plurality of stay points, Among the stay points, the stay point with the most stay days is extracted as a home stay point, and the stay point with the second stay days is the second place among the plurality of stay points. And a behavior feature extraction device having a behavior feature extraction means for extracting as a point.
The behavior feature extraction method according to an aspect of the present invention includes a plurality of residence points including a residence point of each of a plurality of residences of the user, a residence start date and time at the residence point, and a residence end date and time at the residence point. The residence days are calculated for each of the residence points, and the residence point with the largest residence days is extracted as the home residence point among the plurality of residence points, and the residence days is 2 among the plurality of residence points. The second most frequent residence point is extracted as a workplace residence point.
The computer-readable recording medium according to one embodiment of the present invention stores, in a computer, residence point information including each residence point of a plurality of residences of a user, a residence start date and time at the residence point, and a residence end date and time at the residence point. Based on each of the plurality of stay points, the stay days are calculated, and among the plurality of stay points, the stay point having the most stay days is extracted as a home stay point, and among the plurality of stay points, An action feature extraction program for executing a process of extracting the stay point having the second most stay days as a workplace stay point is stored.

The effect of the present invention is that the home position and the work position characterizing the user's behavior can be extracted from the position information history without using environmental information such as illuminance and noise and information on the user's working hours.

It is a block diagram which shows the characteristic structure of 1st embodiment of this invention. It is a block diagram which shows the structure of the action feature extraction system in 1st embodiment of this invention. It is a flowchart which shows the positional information acquisition process of the terminal 100 in 1st embodiment of this invention. It is a flowchart which shows the action type extraction process of the action type extraction apparatus 200 in 1st embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in 1st embodiment of this invention. It is a figure which shows the example of the stay point information in 1st embodiment of this invention. It is a figure which shows the example of the calculation method of the staying days in 1st embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in 1st embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in 1st embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in 2nd embodiment of this invention. It is a figure which shows the example of the stay point information in 2nd embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in 2nd embodiment of this invention. It is a figure which shows the example of the calculation result of the total residence time in 2nd embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in 2nd embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in 3rd embodiment of this invention. It is a figure which shows the example of the stay point information in 3rd embodiment of this invention. It is a figure which shows the example of the calculation method of the staying days in 3rd embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in 3rd embodiment of this invention. It is a figure which shows the example of the calculation result of the night staying days in 3rd embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in 3rd embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in the 4th embodiment of this invention. It is a figure which shows the example of the staying point information in 4th embodiment of this invention. It is a figure which shows the example of the calculation result of the stay days in 4th embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in 4th embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in the 5th embodiment of this invention. It is a figure which shows the example of the stay point information in 5th embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in 5th embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in the 5th Embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in the 6th embodiment of this invention. It is a figure which shows the example of the stay point information in the 6th Embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in the 6th Embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in the 6th Embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 6th Embodiment of this invention. It is a figure which shows the other example of the calculation result of the staying days in the 6th Embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 6th Embodiment of this invention. It is a flowchart which shows the action feature extraction process of the action feature extraction apparatus 300 in the 7th embodiment of this invention. It is a figure which shows the example of the stay point information in the 7th Embodiment of this invention. It is a figure which shows the example of the action characteristic data 311 in the 7th Embodiment of this invention. It is a figure which shows the example of the calculation result of the staying days in the 7th embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 7th Embodiment of this invention. It is a figure which shows the other example of the calculation result of the staying days in the 7th Embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 7th Embodiment of this invention. It is a figure which shows the other example of the calculation result of the staying days in the 7th Embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 7th Embodiment of this invention. It is a figure which shows the other example of the calculation result of the staying days in the 7th Embodiment of this invention. It is a figure which shows the other example of the action characteristic data 311 in the 7th Embodiment of this invention.

First, behavior type information (residence point information) and a living behavior model in the embodiment of the present invention will be described.
In the behavior feature extraction system according to the embodiment of the present invention, the terminal 100 that moves together with the user periodically acquires position information. Then, the behavior type extraction device 200 extracts the stay point information as the user's behavior type information based on the acquired position information. The staying point information is information regarding a staying point (a staying place) where the user has visited and stayed, and includes a staying point identifier and a staying date and time.
Here, the user's life behavior model is defined as follows.
It is considered that a user who has a general social life stays in a specific place for a certain time or more almost every day for sleeping. Therefore, it can be estimated that among the user's stay points, the stay point with the most stay days indicates the user's home.
Further, it is considered that the user stays at a specific place several times a week for a certain time or more for work, school work or the like. There are various time forms such as day shift, night shift, shift work, full time, part time, short-term part-time job, etc., in the embodiment of the present invention. Estimate the staying point with many days as the workplace where the user commute. In addition, when a user is a student, it is good also as a school where a user goes to school at a staying point with the longest staying day after home.
Here, the number of staying days is not counted in units of one day in the calendar, but is counted as one day until 24 hours have elapsed from the time when the staying started in the place for the first time in one day. For example, if you start staying at 9 pm and end your stay the next day at 8 am, the staying end time (8 am the next day) is within 24 hours from the staying start time (9 pm). One day. Also, if you go home at 9:00 pm and leave the house the next day at 10 pm, the staying end time (10 pm the next day) is over 24 hours from the staying start time (9 pm), so the staying days are 2 It will be a day. Moreover, even if it stays at the same staying point many times during 24 hours, the staying days are 1 day. For example, even when staying at the same staying point once in the morning and in the afternoon, the staying day at the staying point is one day.
(First embodiment)
Next, a first embodiment of the present invention will be described.
First, the configuration of the first embodiment of the present invention will be described. FIG. 2 is a block diagram showing a configuration of the behavior feature extraction system according to the first embodiment of the present invention.
Referring to FIG. 2, the behavior feature extraction system includes a terminal 100, a behavior type extraction device 200, a behavior feature extraction device 300, and a behavior feature reference device 400. The terminal 100 and the behavior type extraction device 200, the behavior type extraction device 200 and the behavior feature extraction device 300, and the behavior feature extraction device 300 and the behavior feature reference device 400 are connected by a network (not shown) and can communicate with each other.
The terminal 100 is an information terminal that can move with the user. In the embodiment of the present invention, the terminal 100 is a mobile phone terminal. The terminal 100 may be an information terminal such as a PDA (Personal Data Assistant), a personal computer, or a car navigation system terminal, instead of a mobile phone terminal. Although only one terminal 100 is illustrated in FIG. 2, a plurality of terminals 100 may be provided. The terminal 100 acquires the position information of the terminal 100 and outputs it as position information data 111.
Here, the terminal 100 includes a position information acquisition unit 101. The position information acquisition unit 101 acquires position information of the terminal 100 by GPS (Global Positioning System). The terminal 100 includes an antenna and receives radio waves transmitted from a GPS satellite (not shown). The position information acquisition unit 101 calculates the position (positioning point) of the terminal 100 based on the radio wave received from the satellite. Further, the position information acquisition unit 101 calculates a positioning point and simultaneously acquires a positioning time. The positioning point calculated by the position information acquisition unit 101 is the position of the terminal 100.
The position information acquisition unit 101 acquires the positioning point from the information on the installation position of the reader attached to the reader of an RFID (Radio Frequency IDentification) installed in a specific place (store, etc.), for example, instead of the GPS. May be. Further, the position information acquisition unit 101 may acquire a positioning point by estimating the moving distance of the terminal 100 using an acceleration sensor or a geomagnetic sensor. In addition, the position information acquisition unit 101 may acquire other information related to the positioning point, such as positioning accuracy, at the same time as calculating the positioning point, and may include it in the position information.
The position information acquisition unit 101 periodically calculates a positioning point, and transmits position information including the positioning point, positioning time, and positioning accuracy information to the behavior type extraction device 200 as the position information data 111. The time interval at which position information is acquired by the position information acquisition unit 101 may be set in advance in the position information acquisition unit 101 by the administrator, or may be set in the position information acquisition unit 101 by the user.
The behavior type extraction device 200 extracts the behavior type information of the terminal 100 based on the position information and outputs it as behavior type data 211.
Here, the behavior type extraction apparatus 200 includes a behavior type extraction unit 201 and a behavior type storage unit 202.
The behavior type extraction unit 201 receives the position information data 111 from the terminal 100 and extracts the stay point information as the behavior type information of the terminal 100 based on the received position information data 111. In addition, the behavior type extraction unit 201 stores the extracted stay point information as behavior type data 211 in the behavior type storage unit 202 for each terminal 100. Furthermore, the behavior type extraction unit 201 transmits the behavior type data 211 to the behavior feature extraction device 300.
FIG. 6 is a diagram showing an example of staying point information in the first embodiment of the present invention. The stay point information includes a stay data identifier for identifying each stay, stay point identifiers P1 to P4 for identifying a stay point in each stay, a stay start date and time, and a stay end date and time. The stay point information may include information indicating the position of the stay point.
The behavior feature extraction device 300 extracts behavior feature information of the terminal 100 based on the behavior type information and outputs it as behavior feature data 311.
Here, the behavior feature extraction apparatus 300 includes a behavior type reference unit 301, a behavior feature extraction unit 302, and a behavior feature storage unit 303.
The behavior type reference unit 301 receives the behavior type data 211 from the behavior type extraction device 200.
Based on the behavior type data 211, the behavior feature extraction unit 302 extracts a home residence point that is a residence point that is estimated as a home location and a workplace residence point that is a residence point that is estimated as a workplace location. Information and workplace stay point information are generated. Here, the behavior feature extraction unit 302 calculates the residence days of each residence point based on the behavior type data 211, the residence point with the most residence days is the home residence point, and the residence point with the second residence days is the second. It will be the staying point in the workplace. The behavior feature extraction unit 302 stores behavior feature data 311 including the generated home residence point information and workplace residence point information in the behavior feature storage unit 303 for each terminal 100. Furthermore, the behavior feature extraction device 300 transmits behavior feature data 311 to the behavior feature reference device 400.
FIG. 9 is a diagram showing an example of the behavior feature data 311 in the first embodiment of the present invention. The behavior characteristic data 311 includes home stay point information and workplace stay point information. Home residence point information includes the residence point identifier of the residence point extracted as the home residence point. The workplace residence point information includes the residence point identifier of the residence point extracted as the workplace residence point. The home stay point information may include information indicating the stay date and time at the home stay point and the position of the home stay point. Similarly, the workplace residence point information may include information indicating the residence date and time and the location of the workplace residence point at the workplace residence point.
The behavior feature reference device 400 is a server on which an application that uses the behavior feature information of the user operates. Here, the application may be any application as long as the behavior feature information is used. For example, the application may be an application that provides an advertisement distribution service or the like based on user behavior characteristic information. The application of the behavior feature reference device 400 performs a predetermined process based on the behavior type information received from the behavior feature extraction device 300.
Note that the terminal 100, the behavior type extraction device 200, the behavior feature extraction device 300, and the behavior feature reference device 400 may each be a computer that operates according to a program. In this case, the terminal 100, the behavior type extraction device 200, the behavior feature extraction device 300, and the behavior feature reference device 400 include a storage unit, a processing unit, an input / output unit, and a communication unit (not shown), which are electrically connected by a common bus. Connected. The storage unit includes a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory, and the like, and stores programs and data for realizing the functions of each device. The processing unit is configured by a CPU (Central Processing Unit), and executes the function of each device by reading the program in the storage unit and performing processing. The input / output unit includes an LCD (Liquid Crystal Display), a keyboard, a mouse, a speaker, and the like, and is an input / output interface with an administrator of each device. The communication unit performs communication with other devices by performing wireless communication or wired communication. In this way, the terminal 100, the behavior type extraction device 200, the behavior feature extraction device 300, and the behavior feature reference device 400 are realized.
In the behavior feature extraction system according to the embodiment of the present invention, the behavior feature extraction device 300 is different from the terminal 100, the behavior type extraction device 200, and the behavior feature reference device 400. However, the behavior feature extraction device 300 may constitute one or more of the terminal 100, the behavior type extraction device 200, and the behavior feature reference device 400, and one device. For example, the behavior type extraction device 200 and the behavior feature extraction device 300 may constitute one device. In addition, each component of the behavior feature extraction apparatus 300 may be disposed in a physically different place and connected via a network. That is, the configuration of the behavior feature extraction system illustrated in FIG. 2 is merely an example, and each of the terminal 100, the behavior type extraction device 200, the behavior feature extraction device 300, and the behavior feature reference device 400 includes any component. It can be changed flexibly.
Next, the operation of the behavior feature extraction system in the first embodiment of the present invention will be described.
First, the operation of the terminal 100 in the first embodiment of the present invention will be described. FIG. 3 is a flowchart showing position information acquisition processing of the terminal 100 in the first embodiment of the present invention.
The position information acquisition unit 101 of the terminal 100 receives radio waves from the satellite and periodically calculates positioning points (step S101). Here, the position information acquisition unit 101 acquires the positioning accuracy information and the positioning time at the same time when calculating the positioning point (step S102). The position information acquisition unit 101 transmits position information data 111 including a positioning point, positioning accuracy information, and positioning time to the behavior type extraction device 200 (step S103).
Next, operation | movement of the action type | mold extraction apparatus 200 in 1st embodiment of this invention is demonstrated. FIG. 4 is a flowchart showing the behavior type extraction process of the behavior type extraction device 200 according to the first embodiment of the present invention.
The behavior type extraction unit 201 of the behavior type extraction device 200 receives the position information data 111 from the terminal 100 (step S201). The behavior type extraction unit 201 extracts the stay point information from the position information data 111 transmitted from the terminal 100 (step S202). The behavior type extraction unit 201 stores the extracted stay point information in the behavior type storage unit 202 as behavior type data 211 (step S203). In response to the acquisition request for the behavior type data 211 received from the behavior feature extraction device 300, the behavior type extraction device 200 transmits the behavior type data 211 to the behavior feature extraction device 300 (step S204).
Next, the operation of the behavior feature extraction apparatus 300 according to the first embodiment of the present invention will be described. FIG. 5 is a flowchart showing the behavior feature extraction process of the behavior feature extraction apparatus 300 according to the first embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S301).
For example, the behavior type reference unit 301 receives the stay point information as illustrated in FIG. 6 as the behavior type data 211.
The behavior feature extraction unit 302 calculates the stay days of each stay point included in the stay point information received by the behavior type reference unit 301 (step S302).
FIG. 7 is a diagram showing an example of a method for calculating the staying days in the first embodiment of the present invention. Moreover, FIG. 8 is a figure which shows the example of the calculation result of the staying days in 1st embodiment of this invention. For example, the behavior feature extraction unit 302 calculates the stay days as shown in FIG. 7 for the stay points of the stay point identifiers P1 to P4 included in the stay point information of FIG. As a result, the behavior feature extraction unit 302 obtains the calculation result of the stay days as shown in FIG.
Based on the stay days calculated for each stay point, the behavior feature extraction unit 302 extracts the stay point with the most stay days as a home stay point, and generates home stay point information (step S303). In addition, the behavior feature extraction unit 302 extracts a stay point having the second most stay days as a workplace stay point and generates workplace stay point information (step S304). The behavior feature extraction unit 302 stores home residence point information and workplace residence point information in the behavior feature storage unit 303 (step S305).
For example, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P1 having the most stay days as the stay point at home based on the calculation result of the stay days in FIG. Further, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P2 having the second most stay days as the workplace stay point. The behavior feature extraction unit 302 generates home residence point information and workplace residence point information as illustrated in FIG. 9 and stores them in the behavior feature storage unit 303.
In response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400, the behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 (step S306).
Note that the behavior feature extraction device 300 may periodically execute the processing from step S301 to step S305 at predetermined time intervals, or obtain behavior feature data 311 received from the behavior feature reference device 400. It may be executed in response to a request.
Thereafter, the behavior feature information extracted by the behavior feature extraction unit 302 is used by an application on the behavior feature reference device 400.
Thus, the operation of the first embodiment of the present invention is completed.
Next, a characteristic configuration of the first embodiment of the present invention will be described.
FIG. 1 is a block diagram showing a characteristic configuration of the first embodiment of the present invention.
Referring to FIG. 1, the behavior feature extraction apparatus 300 includes a behavior type storage unit 202 and a behavior feature extraction unit 302. Here, the behavior type storage unit 202 stores stay point information including each stay point of the plurality of stays of the user, a stay start date and time at the stay point, and a stay end date and time at the stay point. The behavior feature extraction unit 302 calculates the residence days for each of the plurality of residence points based on the residence point information in the behavior type storage unit 202, and selects the residence point with the largest residence days among the plurality of residence points. The residence point is extracted as the home residence point, and the residence point having the second largest residence day among the plurality of residence points is extracted as the workplace residence point.
According to the first embodiment of the present invention, the home position and the work position can be extracted from the position information history without using environmental information such as illuminance and noise and information on the user's working hours. The reason is that the behavior feature extraction unit 302 calculates the stay days for each of the plurality of stay points based on the stay point information, and selects the stay point with the largest stay days among the plurality of stay points. This is because the residence point is extracted as a home residence point, and the residence point having the second largest number of residence days among the plurality of residence points is extracted as a workplace residence point.
In addition, according to the first embodiment of the present invention, the home position and the work position can be extracted regardless of the time form in which the user works. The reason is that the behavior feature extraction unit 302 calculates the residence days based on the presence or absence of residence in units of 24 hours with reference to the residence start date and time of the first residence in one day at each residence point, This is because the home position and the workplace position are extracted based on the stay days.
(Second embodiment)
Next, a second embodiment of the present invention will be described.
In the second embodiment of the present invention, when there are a plurality of residence points with the most residence days in the extraction of home residence points and workplace residence points by the behavior feature extraction device 300, the total residence time at each residence point This is different from the first embodiment of the present invention in that the home residence point and the workplace residence point are extracted based on the above.
For many users, it can be estimated that the staying time at home is longer than the staying time at work. Therefore, in the second embodiment of the present invention, when there are a plurality of stay points having the longest stay days, the home stay point and the workplace stay point are extracted based on the total stay time at each stay point.
The configuration of the second embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
Next, operation | movement of the action feature extraction apparatus 300 in 2nd embodiment of this invention is demonstrated. FIG. 10 is a flowchart showing the behavior feature extraction process of the behavior feature extraction apparatus 300 according to the second embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S401).
FIG. 11 is a diagram illustrating an example of staying point information according to the second embodiment of the present invention. For example, the behavior type reference unit 301 receives the stay point information as illustrated in FIG. 11 as the behavior type data 211.
The behavior feature extraction unit 302 calculates the stay days of each stay point included in the stay point information received by the behavior type reference unit 301 (step S402).
FIG. 12 is a diagram illustrating an example of a calculation result of the staying days in the second embodiment of the present invention. For example, the behavior feature extraction unit 302 obtains the calculation result of the stay days as shown in FIG. 12 for the stay points of the stay point identifiers P1 to P4 included in the stay point information of FIG.
Next, the behavior feature extraction unit 302 determines whether or not there are a plurality of stay points having the most stay days based on the calculated stay days of each stay point (step S403). Here, when there is one staying point with the most staying days (step S403 / No), the behavior feature extraction unit 302, like the first embodiment of the present invention (steps S303 and S304), each staying point. Based on the stay days, the home stay point and the workplace stay point are extracted (steps S404 and S405).
On the other hand, when there are a plurality of stay points with the most stay days (step S403 / Yes), the behavior feature extraction unit 302 calculates the total stay time from the stay point information for each stay point with the most stay days (step S406). ).
FIG. 13 is a diagram showing an example of the calculation result of the total residence time in the second embodiment of the present invention. For example, the behavior feature extraction unit 302 calculates the total residence time from the residence start date and time and the residence end date and time of the residence point information of FIG. 11 for the residence point identifiers P1 and P2 of the residence point with the most residence days in FIG. The calculation result of the total residence time as shown in FIG. 13 is obtained.
Based on the calculated total residence time of each residence point, the behavior feature extraction unit 302 extracts the residence point with the longest total residence time as the home residence point and generates home residence point information (step S407). In addition, the behavior feature extraction unit 302 extracts a stay point having the second longest stay time as a workplace stay point, and generates workplace stay point information (step S408).
FIG. 14 is a diagram showing an example of the behavior feature data 311 in the second embodiment of the present invention. For example, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P1 having the longest total stay time as the home stay point based on the calculation result of the total stay time in FIG. Further, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P2 having the second longest stay time as the workplace stay point. Then, the behavior feature extraction unit 302 generates home residence point information and workplace residence point information as shown in FIG.
The behavior feature extraction unit 302 stores the home residence point information and the workplace residence point information in the behavior feature storage unit 303 (step S409). The behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 in response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400 (step S410).
Thus, the operation of the second embodiment of the present invention is completed.
According to the second embodiment of the present invention, the home stay point and the workplace stay point can be extracted even when there are a plurality of stay points with the longest stay days. The reason is that when there are a plurality of stay points having the most stay days, the behavior feature extraction unit 302 extracts the home stay point and the workplace stay point based on the total stay time at each stay point.
(Third embodiment)
Next, a third embodiment of the present invention will be described.
In the third embodiment of the present invention, in the extraction of the home stay point and the workplace stay point by the behavior feature extraction device 300, when the number of positioning days in the stay point information is equal to or less than a predetermined number of days, the night stay days are used. It differs from the first embodiment of the present invention in that the home residence point is extracted.
For many users, when positioning is started from Monday, the stay days at the home stay point and the work stay point are the same until the first weekend comes. Here, if you work overtime for a long time at work, or if you work for a long time at a place other than your home or work, the residence time at the workplace will become longer or the residence time at your home will become shorter. Sometimes. In this case, as in the second embodiment of the present invention, even if the total residence time at each residence point is used, the home residence point and the workplace residence point may not be correctly extracted. Therefore, in the third embodiment of the present invention, when the number of positioning days in the stay point information is equal to or less than the predetermined number of days, the home stay point is determined based on the night stay days calculated for the stay including the predetermined time at night. Extract.
The configuration of the third embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
Next, the operation of the behavior feature extraction apparatus 300 according to the third embodiment of the present invention will be described. FIG. 15 is a flowchart showing behavior feature extraction processing of the behavior feature extraction apparatus 300 according to the third embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S501).
FIG. 16 is a diagram showing an example of staying point information according to the third embodiment of the present invention. For example, the behavior type reference unit 301 receives the stay point information as illustrated in FIG. 16 as the behavior type data 211.
The behavior feature extraction unit 302 calculates the stay days of each stay point included in the stay point information received by the behavior type reference unit 301 (step S502).
FIG. 17 is a diagram illustrating an example of a method for calculating the staying days in the third embodiment of the present invention. Moreover, FIG. 18 is a figure which shows the example of the calculation result of the staying days in 3rd embodiment of this invention. For example, the behavior feature extraction unit 302 calculates the stay days as shown in FIG. 17 for the stay points of the stay point identifiers P1 to P3 included in the stay point information of FIG. As a result, the behavior feature extraction unit 302 obtains the calculation result of the stay days as shown in FIG.
Next, the behavior feature extraction unit 302 determines that the number of days of positioning of all staying point information to be extracted in step S502 (the number of days from the first staying start time to the last staying end time included in the staying point information) is a predetermined number of days. It is determined whether it is below (step S503). Here, when the number of positioning days exceeds the predetermined number of days (step S503 / No), the behavior feature extraction unit 302, as in the first embodiment of the present invention (steps S303 and S304), Based on the stay days, the home stay point and the workplace stay point are extracted (steps S504 and S505).
On the other hand, when the number of positioning days is equal to or less than the predetermined number of days (step S503 / Yes), the behavior feature extraction unit 302 calculates the number of staying days at night at each staying point (step S506). Here, the number of staying days at night is the number of staying days calculated for staying including a predetermined time at night.
FIG. 19 is a diagram illustrating an example of a calculation result of the number of days staying at night in the third embodiment of the present invention. For example, when the number of days of positioning is 7 days (predetermined number of days) or less, and the number of days of staying at night is calculated, the number of days of positioning in the staying point information in FIG. 16 is 3 days. The number of staying days at night is calculated for the staying points of the staying point identifiers P1 to P3 included in the information. Here, assuming that the staying days at night is the staying days calculated for staying including 3:00 am (predetermined time), the behavior feature extraction unit 302 calculates the staying days at night as shown in FIG. As a result, the behavior feature extraction unit 302 obtains the calculation result of the staying days at night as shown in FIG.
Based on the calculated number of staying days at each staying point, the behavior feature extraction unit 302 extracts the staying point having the largest number of staying days at night as a home staying point, and generates home staying point information (step S507). In addition, the behavior feature extraction unit 302 extracts the stay point with the most stay days as the work place stay point based on the calculated stay days of each stay point, and generates work place stay point information. (Step S508).
FIG. 20 is a diagram illustrating an example of the behavior feature data 311 according to the third embodiment of this invention. For example, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P1 having the largest number of stay days at night as the stay point at home based on the calculation result of the stay days at night in FIG. Further, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P2 having the most stay days as a work stay point, excluding the stay point at home, based on the calculation result of the stay days in FIG.
The behavior feature extraction unit 302 stores the home stay point information and the workplace stay point information in the behavior feature storage unit 303 (step S509). The behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 in response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400 (step S510).
Thus, the operation of the third embodiment of the present invention is completed.
According to the third embodiment of the present invention, there are a plurality of residence points with the most residence days, and even when the total residence time at the workplace residence point is longer than the total residence time at the home residence point, And workplace residence points can be extracted. The reason is that when the number of positioning days in the stay point information is equal to or less than the predetermined number of days, the behavior feature extraction unit 302 extracts the home stay point based on the night stay days calculated for the stay including the predetermined time at night. Because.
(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described.
In the fourth embodiment of the present invention, in the extraction of home stay points and workplace stay points by the behavior feature extraction device 300, the stay days are calculated for stay points staying for a predetermined time or more per day. This is different from the first embodiment of the present invention.
The number of days staying at the nearest station or transfer station of the home or work place where the user stays for a short time every day or every working day can be close to the stay time at the home stay point or work stay point. High nature. Therefore, in the fourth embodiment of the present invention, in order to prevent these short-time stay points from being extracted as homes or workplaces, the stay points staying for a predetermined time or more per day are targeted. Calculate the number of days.
The configuration of the fourth embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
Next, the operation of the behavior feature extraction apparatus 300 according to the fourth embodiment of the present invention will be described. FIG. 21 is a flowchart showing a behavior feature extraction process of the behavior feature extraction apparatus 300 according to the fourth embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S601).
FIG. 22 is a diagram illustrating an example of staying point information according to the fourth embodiment of the present invention. For example, the behavior type reference unit 301 receives the stay point information as illustrated in FIG. 22 as the behavior type data 211.
The behavior feature extraction unit 302 extracts a stay point that stays for a predetermined time or more per day from the stay point information received by the behavior type reference unit 301 (step S602).
For example, when extracting a stay point that has stayed for 2 hours (predetermined time) per day, the behavior feature extraction unit 302 extracts the stay points of the stay point identifiers P1 and P2 from the stay point information of FIG. To do.
Similar to the first embodiment of the present invention (steps S302 to S306), the behavior feature extraction unit 302 uses the staying points extracted in step S602 as a target, based on the staying days of each staying point. The stay point and the workplace stay point are extracted, stored in the behavior feature storage unit 303, and transmitted to the behavior feature reference device 400 (steps S603 to S607).
FIG. 23 is a diagram illustrating an example of the calculation result of the stay days in the fourth embodiment of the present invention. For example, the behavior feature extraction unit 302 calculates the stay days for the stay points of the stay point identifiers P1 and P2, and obtains the stay day calculation result as shown in FIG.
FIG. 24 is a diagram showing an example of the behavior feature data 311 in the fourth embodiment of the present invention. For example, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P1 with the most stay days as the home stay point based on the calculation result of the stay days in FIG. Further, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P2 having the second most stay days as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
Thus, the operation of the fourth embodiment of the present invention is completed.
In addition, in the fourth embodiment of the present invention, the residence days of each residence point are targeted for residence points that stay for a predetermined time or more per day, as in the first embodiment of the present invention. The home residence point and the workplace residence point were calculated, and the residence days that stayed for a predetermined time or more per day are the same as in the second embodiment of the present invention. In the case where there are a plurality of stay points with a large number of times, the total stay time at each stay point may be calculated to extract the home stay point and the workplace stay point. In addition, for the staying points staying for a predetermined time or more per day, as in the third embodiment of the present invention, when the number of positioning days in the staying point information is equal to or less than the predetermined number of days, the number of staying days at night The home residence point may be extracted based on.
Thus, the operation of the fourth embodiment of the present invention is completed.
According to the fourth embodiment of the present invention, it is possible to extract the home residence point and the workplace residence point even when there is a residence point other than the home or workplace that stays every day or every working day. The reason is that the behavior feature extraction unit 302 calculates the stay days for the stay points where the stay time per day is equal to or longer than a predetermined time.
(Fifth embodiment)
Next, a fifth embodiment of the present invention will be described.
In the fifth embodiment of the present invention, in the extraction of home stay points and workplace stay points by the behavior feature extraction device 300, the present invention is characterized in that the stay days are calculated for stay point information for a predetermined period. This is different from the first embodiment.
When the home or workplace is moved, the staying point at the relocation destination cannot be promptly extracted as the home staying point or the work staying point because the staying days at the relocation destination of the home or work are few immediately after the relocation. Therefore, in the fifth embodiment of the present invention, the staying days are calculated for the staying point information for a predetermined period.
The configuration of the fifth embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
Next, the operation of the behavior feature extraction apparatus 300 according to the fifth embodiment of the present invention will be described. FIG. 25 is a flowchart showing a behavior feature extraction process of the behavior feature extraction apparatus 300 according to the fifth embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S701).
FIG. 26 is a diagram illustrating an example of staying point information according to the fifth embodiment of the present invention. For example, the behavior type reference unit 301 receives the stay point information as illustrated in FIG. 26 as the behavior type data 211 at the current date and time “2010/02/10 17:30”.
The behavior feature extraction unit 302 extracts the residence point information for a predetermined period (fixed period) from the current date and time from the residence point information received by the behavior type reference unit 301 (step S702). In addition, in step S701, the behavior feature extraction unit 302 may request and receive the stay point information for a predetermined period from the current date and time to the behavior type extraction device 200.
For example, when the stay point information from the current date and time to 7 days before (predetermined period) is extracted, the behavior feature extraction unit 302 uses the stay start time “2010/02/03 19: The staying point information after “00” is extracted.
Similar to the first embodiment of the present invention (steps S302 to S306), the behavior feature extraction unit 302 targets the staying point information extracted in step S702, based on the staying days of each staying point. The home residence point and the workplace residence point are extracted, stored in the behavior feature storage unit 303, and transmitted to the behavior feature reference device 400 (steps S703 to S707).
FIG. 27 is a diagram illustrating an example of the calculation result of the staying days in the fifth embodiment of the present invention. For example, the behavior feature extraction unit 302 obtains the calculation result of the stay days as shown in FIG. 27 for the stay point information after the stay start time “2010/02/03 19:00”.
FIG. 28 is a diagram showing an example of behavior feature data 311 in the fifth embodiment of the present invention. For example, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P1 having the most stay days as the stay point at home based on the calculation result of the stay days in FIG. In addition, the behavior feature extraction unit 302 extracts the stay point of the stay point identifier P6 having the second most stay days as the workplace stay point. The behavior feature extraction unit 302 generates home residence point information and workplace residence point information as shown in FIG.
Thus, the operation of the fifth embodiment of the present invention is completed.
In the fifth embodiment of the present invention, the residence days at each residence point are calculated for residence point information for a predetermined period, as in the first embodiment of the present invention, Points and workplace stay points are extracted, but when there are a plurality of stay points with the most stay days, as in the second embodiment of the present invention, for stay point information for a predetermined period, The total residence time at the residence point may be calculated to extract the home residence point and the workplace residence point. In addition, for the stay point information for a predetermined period, as in the third embodiment of the present invention, when the number of positioning days in the stay point information is equal to or less than the predetermined number of days, the home stay point is based on the stay days at night. May be extracted.
Thus, the operation of the fifth embodiment of the present invention is completed.
According to the fifth embodiment of the present invention, even if the home or the workplace is moved, the moved home residence point or the workplace residence point can be quickly extracted. The reason is that the behavior feature extraction unit 302 extracts the staying point and the staying point in the workplace using the staying point information on staying performed for a predetermined period from the current date and time.
(Sixth embodiment)
Next, a sixth embodiment of the present invention will be described.
In the sixth embodiment of the present invention, in the extraction of the home stay point and the work place stay point by the behavior feature extraction device 300, the stay at the home stay point or the work stay point previously extracted is not performed for a predetermined period. In this case, it differs from the first embodiment of the present invention in that the home stay point and the workplace stay point are extracted for the stay point information after the date and time when the stay is not performed.
The configuration of the sixth embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
31, FIG. 33 and FIG. 35 are diagrams showing examples of behavior feature data 311 in the sixth embodiment of the present invention. The behavior feature data 311 includes stay point analysis start point information, home stay point information, and workplace stay point information. Here, the stay point analysis start point information includes a stay data identifier indicating the start point of the stay point information used when extracting the home stay point and the workplace stay point.
Next, the operation of the behavior feature extraction device 300 according to the sixth embodiment of the present invention will be described. FIG. 29 is a flowchart showing a behavior feature extraction process of the behavior feature extraction apparatus 300 according to the sixth embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S801).
The behavior feature extraction unit 302 refers to the behavior feature data 311 stored in the behavior feature storage unit 303 and the received stay point information, and extracts the previous behavior feature from the current date and time for a predetermined period (a certain period). It is determined whether or not the stay at the home stay point and the work stay point extracted in the process has been performed (step S802). Here, the behavior feature extraction unit 302 adds the residence point of the home residence point information included in the behavior feature data 311 stored in the behavior feature storage unit 303 to the residence point information for a predetermined period (fixed period) from the current date and time. It is determined whether the identifier and the stay point identifier of the workplace stay point information exist.
When staying at the home stay point and the work place stay point previously extracted has been performed for a predetermined period from the current date and time (step S802 / Yes), the behavior feature extraction unit 302 proceeds to the processing after step S804.
On the other hand, if the stay at the home stay point or the work stay point extracted last time has not been performed for a predetermined period from the current date and time (step S802 / No), the behavior feature extraction unit 302 stores the stay point analysis start point information. Is updated with the staying data identifier of the staying performed after the last staying at the home staying point or workplace staying point where the staying is not performed, and stored in the behavior feature storage unit 303 (step S803).
The behavior feature extraction unit 302 extracts the stay point information after the stay data identifier indicated by the stay point analysis start point information from the stay point information received by the behavior type reference unit 301 (step S804). Similar to the first embodiment of the present invention (steps S302 to S304), the behavior feature extraction unit 302 targets the stay point information extracted in step S803 based on the stay days of each stay point. Home residence points and workplace residence points are extracted (steps S805 to S807).
FIG. 30 is a diagram illustrating an example of staying point information according to the sixth embodiment of the present invention. 32 and 34 are diagrams illustrating examples of calculation results of the staying days in the sixth embodiment of the present invention.
For example, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 14 in FIG. 30 as the behavior type data 211 at the current date and time “2010/02/09 17:30”. In the previous behavior feature extraction process, the behavior feature extraction unit 302 extracts home residence points and workplace residence points based on the residence point information (retention data identifiers 1 to 12). Action feature data 311 in FIG. 31 is stored.
Here, when it is determined whether the stay at the home stay point and the work stay point extracted last time is performed from the current date and time to 5 days ago (predetermined period), the current date “2010/02/09” is determined. 17:30 ”to 5 days ago, because staying at the home stay point (stay point identifier P1) and workplace stay point (stay point identifier P2) extracted last time is performed (step S802 / Yes). The behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 1 to 14 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days for the stay point information of the stay data identifiers 1 to 14 as shown in FIG. 32, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P2 with the second most stay days is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
Next, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 16 in FIG. 30 as the behavior type data 211 at the current date and time “2010/02/10 17:30”.
Here, since the stay at the current workplace stay point (stay point identifier P2) is not performed between the current date and time “2010/02/10 17:30” and 5 days ago (step S802 / No), the action The feature extraction unit 302 updates the stay point analysis start point information with the stay data identifier 11 of the stay performed after the last stay (stay data identifier 10) at the work stay point (stay point identifier P2). The behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 11 to 16 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days as shown in FIG. 34 for the stay point information of the stay data identifiers 11 to 16, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P7 having the second stay day is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
The behavior feature extraction unit 302 stores the home residence point information and the workplace residence point information in the behavior feature storage unit 303 (step S808). In response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400, the behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 (step S809).
Thus, the operation of the sixth embodiment of the present invention is completed.
According to the sixth embodiment of the present invention, even if the home or the workplace is moved, the moved home residence point or the workplace residence point can be quickly extracted. The reason is that if the behavior feature extraction device 300 has not stayed at the home stay point or the work stay point extracted last time for a predetermined period, the stay point analysis start point information is displayed after the last stay at the work stay point. To update the home stay point and the workplace stay point using the stay point information after the stay indicated by the stay data identifier included in the stay point analysis start point information. It is.
Further, according to the sixth embodiment of the present invention, it is possible to extract the home stay point and the work place stay point more accurately than in the fifth embodiment of the present invention. The reason for this is that the behavior feature extraction device 300 does not update the stay point analysis start point information when staying at the home stay point or the work stay point previously extracted is performed for a predetermined period. Or, unless the workplace residence point is moved, the residence residence point and the workplace residence point can be extracted using residence point information for as long a period as possible.
(Seventh embodiment)
Next, a seventh embodiment of the present invention will be described.
In the seventh embodiment of the present invention, in the extraction of the home residence point and the workplace residence point by the behavior feature extraction device 300, the home residence point and the workplace residence point extracted when the residence point analysis start point information is updated. Is the same as the home stay point and the workplace stay point extracted before updating the last stay point analysis start point information, the stay point analysis start point information is updated before the last stay point analysis start point information is updated. This is different from the sixth embodiment of the present invention in that it returns to the stay point analysis start point information used in the above.
The configuration of the seventh embodiment of the present invention is the same as the configuration of the first embodiment of the present invention.
38, 40, 42, 44, and 46 are diagrams showing examples of behavior feature data 311 in the seventh embodiment of the present invention. The behavior characteristic data 311 includes stay point analysis start point information, home stay point information, and workplace stay point information, pre-update stay point analysis start point information, pre-update home stay point information, and pre-update workplace stay point information. Including. Here, the stay point analysis start point information before update, home stay point information before update, and workplace stay point information before update are the start of the stay point analysis used before updating the previous stay point analysis start point information, respectively. Point information, home stay point information extracted before updating the previous stay point analysis start point information, and workplace stay point information extracted before updating the previous stay point analysis start point information.
Next, the operation of the behavior feature extraction apparatus 300 in the seventh embodiment of the present invention will be described. FIG. 36 is a flowchart showing behavior feature extraction processing of the behavior feature extraction apparatus 300 according to the seventh embodiment of the present invention.
First, the behavior type reference unit 301 of the behavior feature extraction device 300 receives the behavior type data 211 from the behavior type extraction device 200 (step S901).
The behavior feature extraction unit 302 refers to the behavior feature data 311 stored in the behavior feature storage unit 303 and the received stay point information, and extracts the previous behavior feature from the current date and time for a predetermined period (a certain period). It is determined whether or not staying at the home staying point and workplace staying point extracted in the process has been performed (step S902). Here, the behavior feature extraction unit 302 adds the residence point of the home residence point information included in the behavior feature data 311 stored in the behavior feature storage unit 303 to the residence point information for a predetermined period (fixed period) from the current date and time. It is determined whether the identifier and the stay point identifier of the workplace stay point information exist.
When staying at the home stay point and workplace stay point extracted last time has been performed for a predetermined period from the current date and time (step S902 / Yes), the same as steps S804 to S808 of the sixth embodiment of the present invention In addition, the home stay point and the workplace stay point are extracted and stored for the stay point information after the stay data identifier indicated by the stay point analysis start point information (steps S903 to S907).
On the other hand, if the stay at the home stay point or the work place stay point previously extracted has not been performed for a predetermined period from the current date and time (step S902 / No), the behavior feature extraction unit 302 stores the stay point analysis start point information. Is updated with the staying data identifier of staying performed after the last staying at the home staying point or workplace staying point where the staying is not performed, and stored in the behavior feature storage unit 303 (step S908). Then, similarly to steps S804 to S808 of the sixth embodiment of the present invention, the home stay point and the workplace stay point are extracted for the stay point information after the stay data identifier indicated by the stay point analysis start point information. And save (steps S908 to S913).
Next, the behavior feature extraction unit 302 refers to the behavior feature data 311, and the extracted residence point identifier of the home residence point and the residence point identifier of the workplace residence point are obtained as follows. It is determined whether or not the stay point identifiers of the stay points coincide with each other (step S914).
If the residence point identifier of the extracted home residence point and the residence point identifier of the workplace residence point do not match the residence point identifier of the home residence point before update and the residence point identifier of the workplace residence point before update (the residence point identifier of the home residence point) Does not match the stay point identifier of the pre-update workplace stay point, or the stay point identifier of the work place stay point does not match the stay point identifier of the pre-update work place stay point (step S914 / No), the behavior feature extraction unit 302 Is the stay point analysis start point information, home stay point information, and workplace stay point information in the previous feature extraction process (before updating the stay point analysis start point information). The stay point information and the pre-update workplace stay point information are set and stored in the behavior feature storage unit 303 (step S915).
On the other hand, when the residence point identifier of the extracted home residence point and the residence point identifier of the workplace residence point match the residence point identifier of the home residence point before update and the residence point identifier of the workplace residence point before update (residence of home residence point) When the point identifier matches the stay point identifier of the pre-update workplace stay point and the stay point identifier of the work place stay point matches the stay point identifier of the pre-update workplace stay point) (step S914 / Yes), behavior feature extraction The unit 302 refers to the behavior feature data 311 and sets the stay data identifier of the stay point analysis start point information before update in the stay point analysis start point information (step S916). The behavior feature extraction unit 302 initializes the pre-update stay point analysis start point information, the pre-update home stay point information, and the pre-update workplace stay point information (sets null), and saves it in the behavior feature storage unit 303 (step) S917).
In response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400, the behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 (step S918).
FIG. 37 is a diagram illustrating an example of staying point information according to the seventh embodiment of the present invention. FIG. 39, FIG. 41, FIG. 43, and FIG. 45 are diagrams showing examples of calculation results of stay days in the seventh embodiment of the present invention.
For example, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 14 in FIG. 37 as the behavior type data 211 at the current date and time “2010/02/09 17:30”. In the previous behavior feature extraction process, the behavior feature extraction unit 302 extracts home residence points and workplace residence points based on the residence point information (retention data identifiers 1 to 12). The behavior characteristic data 311 of FIG. 38 is stored.
Here, when it is determined whether the stay at the home stay point and the work stay point extracted last time is performed from the current date and time to 5 days ago (predetermined period), the current date “2010/02/09” is determined. 17:30 "to 5 days ago, because staying at the home staying point (staying point identifier P1) and workplace staying point (staying point identifier P2) extracted last time is performed (step S902 / Yes). The behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 1 to 14 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days as shown in FIG. 39 for the stay point information of the stay data identifiers 1 to 14, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P2 with the second most stay days is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
Next, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 16 in FIG. 37 as the behavior type data 211 at the current date and time “2010/02/10 17:30”.
Here, the stay at the current workplace stay point (stay point identifier P2) is not performed between the current date and time “2010/02/10 17:30” and 5 days ago (step S902 / No). The feature extraction unit 302 updates the stay point analysis start point information with the stay data identifier 11 of the stay performed after the last stay (stay data identifier 10) at the work stay point (stay point identifier P2). The behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 11 to 16 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days for the stay point information of the stay data identifiers 11 to 16 as shown in FIG. 41, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P7 having the second stay day is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
Here, because the stay point identifier P7 of the extracted workplace stay point does not match the stay point identifier (null) of the pre-update workplace stay point (step S914 / No), the behavior feature extraction unit 302 performs the previous stay point analysis. As shown in FIG. 42, the staying data identifier 1 of the starting point information, the staying point identifier P1 of the home staying point information, and the staying point identifier P2 of the working staying point information are set as shown in FIG. Set to stay point information and pre-update workplace stay point information.
Next, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 26 in FIG. 37 as the behavior type data 211 at the current date “2010/02/17 17:30”.
Here, since the stay at the current workplace stay point (stay point identifier P7) is not performed between the current date and time “2010/02/17 17:30” and 5 days ago (step S902 / No), the action The feature extraction unit 302 updates the stay point analysis start point information with the stay data identifier 21 of the stay performed after the last stay (stay data identifier 20) at the work stay point (stay point identifier P7). The behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 21 to 26 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days as shown in FIG. 43 for the stay point information of the stay data identifiers 21 to 26, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P2 with the second most stay days is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
Here, the residence point identifier P1 of the extracted home residence point coincides with the residence point identifier P1 of the home residence point before update, and the residence point identifier P2 of the extracted workplace residence point is the residence of the workplace residence point before update. Since it coincides with the point identifier P2 (step S914 / Yes), the behavior feature extraction unit 302 sets the stay data identifier 1 of the stay point analysis start point information before update in the stay point analysis start point information as shown in FIG. To do. Then, the behavior feature extraction unit 302 initializes the pre-update stay point analysis start point information, the pre-update home stay point information, and the pre-update workplace stay point information as shown in FIG.
Further, the behavior type reference unit 301 receives the stay point information of the stay data identifiers 1 to 30 in FIG. 37 as the behavior type data 211 at the current date and time “2010/02/19 17:30”.
Here, during the period from the current date and time “2010/02/19 17:30” to 5 days ago, the stay at the home stay point (stay point identifier P1) and workplace stay point (stay point identifier P2) extracted last time is performed. (Step S902 / Yes), the behavior feature extraction unit 302 extracts the stay point information of the stay data identifiers 1 to 30 from the stay point information of FIG.
The behavior feature extraction unit 302 calculates the stay days for the stay point information of the stay data identifiers 1 to 30 as shown in FIG. 45, and sets the stay point of the stay point identifier P1 with the most stay days as the home stay point, 2 The stay point of the stay point identifier P2 with the second most stay days is extracted as the workplace stay point. The behavior feature extraction unit 302 generates home stay point information and workplace stay point information as shown in FIG.
In response to the acquisition request for the behavior feature data 311 received from the behavior feature reference device 400, the behavior feature extraction unit 302 transmits the behavior feature data 311 to the behavior feature reference device 400 (step S918).
Thus, the operation of the seventh embodiment of the present invention is completed.
According to the seventh embodiment of the present invention, it is possible to extract the home residence point and the workplace residence point more accurately than in the sixth embodiment of the present invention. The reason is that the home residence point and workplace residence point extracted when the residence point analysis start point information is updated are the same as the home residence point and workplace residence point extracted before the update of the previous residence point analysis start point information. This is because the behavior feature extraction device 300 returns the stay point analysis start point information to the pre-update stay point analysis start point information used before the update of the previous stay point analysis start point information. Thereby, it is prevented that the period of the residence point information used for extracting the home residence point and the workplace residence point due to the temporary home residence point or the movement of the workplace residence point is shortened.
Although the present invention has been described with reference to the embodiment, the present invention is not limited to the above embodiment. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.
For example, in the above embodiment, the behavior type extraction device 200 extracts the behavior type information based on the position information acquired by the terminal 100, and the behavior feature extraction device 300 extracts the behavior feature information based on the behavior type information. It is assumed that However, the behavior feature extraction device 300 may extract behavior feature information based on the behavior type information manually input to the behavior type extraction device 200.
Moreover, you may combine some structures and operation | movement of the said embodiment. For example, by combining the second embodiment and the third embodiment of the present invention, if the number of positioning days in the stay point information is less than or equal to a predetermined number of days, the home stay point is extracted based on the night stay days. Even when the number of positioning days exceeds a predetermined number of days, if there are multiple residence points with the most residence days, the home residence point and the workplace residence point can be extracted based on the total residence time at each residence point. Good.
This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2010-034180 for which it applied on February 19, 2010, and takes in those the indications of all here.
A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.
(Appendix 1)
For each of the plurality of stays of the user, behavior type storage means for storing stay point information including a stay point, a stay start date and time at the stay point, and a stay end date and time at the stay point;
Based on the stay point information of the behavior type storage means, the stay days are calculated for each of the plurality of stay points, and among the plurality of stay points, the stay point having the most stay days is set as a home stay point. Action feature extraction means for extracting and extracting, as a workplace residence point, the residence point having the second largest residence day among the plurality of residence points;
A behavior feature extraction apparatus comprising:
(Appendix 2)
The behavior feature extraction unit extracts the home residence point and the workplace residence point using the residence point information on the residence performed during a predetermined period until the current date and time (Appendix 1). Feature extraction device.
(Appendix 3)
Based on residence point information including each residence point of the plurality of residences of the user, residence start date and time at the residence point, and residence end date and time at the residence point, the residence days are calculated for each of the plurality of residence points, Among the plurality of staying points, the staying point having the most staying days is extracted as a home staying point, and among the plurality of staying points, the staying point having the second most staying days is extracted as a workplace staying point. Do
Behavior feature extraction method.
(Appendix 4)
In the calculation of the stay days, the stay days based on the presence / absence of the stay every 24 hours based on the stay start date and time of the first stay in each of the plurality of stay points. Calculate
(Supplementary note 3) The behavior feature extraction method according to the description.
(Appendix 5)
In extraction of the home stay point and the workplace stay point, among the plurality of stay points, when there are a plurality of stay points having the most stay days, each of the stay points having the most stay days Calculating the total residence time, and extracting the residence point having the longest total residence time as the home residence point among the residence points having the largest residence days, and out of the residence points having the largest residence days. , And the second residence point having the second total residence time is extracted as the workplace residence point.
The behavior feature extraction method according to (Appendix 3) or (Appendix 4).
(Appendix 6)
In the extraction of the home stay point and the workplace stay point, when the number of days included in the stay point information is equal to or less than a predetermined number of days, the stay days of stay including a predetermined time at night for each of the plurality of stay points The number of staying days at night is calculated, and the staying point having the largest number of staying days at night is extracted as the home staying point among the plurality of staying points, and the home staying point is excluded from the plurality of staying points. The stay point with the most stay days is extracted as the workplace stay point.
The behavior feature extraction method according to any one of (Appendix 3) to (Appendix 5).
(Appendix 7)
In the calculation of the staying days, a staying time per day is calculated for each of the plurality of staying points, and among the plurality of staying points, the staying points having a staying time per day of a predetermined time or more are calculated. Calculate the stay days for
The behavior feature extraction method according to any one of (Appendix 3) to (Appendix 6).
(Appendix 8)
The stay point information further includes a stay data identifier for each of the plurality of stays,
In calculating the stay days,
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time, the residence point analysis start point information is the last of the home residence point or the workplace residence point. Update with the stay data identifier of the stay performed after the stay,
The stay days are calculated for the stay point information after the stay indicated by the stay data identifier included in the stay point analysis start point information.
The behavior feature extraction method according to any one of (Appendix 3) to (Appendix 7).
(Appendix 9)
In extracting the home residence point and the workplace residence point,
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time,
If the home residence point extracted this time does not match the home residence point before update, or if the workplace residence point extracted this time does not match the workplace residence point before update, the residence point analysis start point information in the previous extraction The previously extracted home residence point extracted previously and the workplace residence point extracted last time are set as the before-update residence point analysis start point information, the home residence point before update, and the workplace residence point before update, respectively.
If the home residence point extracted this time matches the home residence point before update, and the workplace residence point extracted this time matches the workplace residence point before update, the residence point analysis start point information before the update Set starting point analysis start point information
(Supplementary note 8) The behavior feature extraction method according to the description.
(Appendix 10)
In the calculation of the staying days, the staying days are calculated using the staying point information on the staying performed during a predetermined period until the current date and time.
The behavior feature extraction method according to any one of (Appendix 3) to (Appendix 7).
(Appendix 11)
On the computer,
Based on residence point information including each residence point of the plurality of residences of the user, residence start date and time at the residence point, and residence end date and time at the residence point, the residence days are calculated for each of the plurality of residence points, Among the plurality of staying points, the staying point having the most staying days is extracted as a home staying point, and among the plurality of staying points, the staying point having the second most staying days is extracted as a workplace staying point. Do
A computer-readable recording medium storing an action feature extraction program for executing processing.
(Appendix 12)
In the calculation of the stay days, the stay days based on the presence / absence of the stay every 24 hours based on the stay start date and time of the first stay in each of the plurality of stay points. Calculate
(Supplementary note 11) A computer-readable recording medium storing the behavior feature extraction program according to (11).
(Appendix 13)
In extraction of the home stay point and the workplace stay point, among the plurality of stay points, when there are a plurality of stay points having the most stay days, each of the stay points having the most stay days Calculating the total residence time, and extracting the residence point having the longest total residence time as the home residence point among the residence points having the largest residence days, and out of the residence points having the largest residence days. , And the second residence point having the second total residence time is extracted as the workplace residence point.
A computer-readable recording medium storing the behavior feature extraction program according to (Appendix 11) or (Appendix 12).
(Appendix 14)
In the extraction of the home stay point and the workplace stay point, when the number of days included in the stay point information is equal to or less than a predetermined number of days, the stay days of stay including a predetermined time at night for each of the plurality of stay points The number of staying days at night is calculated, and the staying point having the largest number of staying days at night is extracted as the home staying point among the plurality of staying points, and the home staying point is excluded from the plurality of staying points. The stay point with the most stay days is extracted as the workplace stay point.
A computer-readable recording medium storing the behavior feature extraction program according to any one of (Appendix 11) to (Appendix 13).
(Appendix 15)
In the calculation of the staying days, a staying time per day is calculated for each of the plurality of staying points, and among the plurality of staying points, the staying points having a staying time per day of a predetermined time or more are calculated. Calculate the stay days for
A computer-readable recording medium storing the behavior feature extraction program according to any one of (Appendix 11) to (Appendix 14).
(Appendix 16)
The stay point information further includes a stay data identifier for each of the plurality of stays,
In calculating the stay days,
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time, the residence point analysis start point information is the last of the home residence point or the workplace residence point. Update with the stay data identifier of the stay performed after the stay,
The stay days are calculated for the stay point information after the stay indicated by the stay data identifier included in the stay point analysis start point information.
A computer-readable recording medium storing the behavior feature extraction program according to any one of (Appendix 11) to (Appendix 15).
(Appendix 17)
In extracting the home residence point and the workplace residence point,
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time,
If the home residence point extracted this time does not match the home residence point before update, or if the workplace residence point extracted this time does not match the workplace residence point before update, the residence point analysis start point information in the previous extraction The previously extracted home residence point extracted previously and the workplace residence point extracted last time are set as the before-update residence point analysis start point information, the home residence point before update, and the workplace residence point before update, respectively.
If the home residence point extracted this time matches the home residence point before update, and the workplace residence point extracted this time matches the workplace residence point before update, the residence point analysis start point information before the update Set starting point analysis start point information
(Supplementary note 16) A computer-readable recording medium storing the behavior feature extraction program according to (16).
(Appendix 18)
In the calculation of the staying days, the staying days are calculated using the staying point information on the staying performed during a predetermined period until the current date and time.
A computer-readable recording medium storing the behavior feature extraction program according to any one of (Appendix 11) to (Appendix 15).

The present invention can be applied not only to information distribution using user behavior characteristic information, but also to user probe data creation in a trade area survey and traffic volume survey.

DESCRIPTION OF SYMBOLS 100 Terminal 101 Position information acquisition part 111 Position information data 200 Behavior type extraction apparatus 201 Behavior type extraction part 202 Behavior type storage part 211 Behavior type data 300 Behavior type extraction apparatus 301 Behavior type reference part 302 Behavior feature extraction part 303 Behavior feature storage part 311 Action Feature Data 400 Action Feature Reference Device

Claims

Behavior type storage means for storing residence point information including each residence point of a plurality of residences of the user, residence start date and time at the residence point, and residence end date and time at the residence point;
Based on the stay point information of the behavior type storage means, the stay days are calculated for each of the plurality of stay points, and among the plurality of stay points, the stay point having the most stay days is set as a home stay point. A behavior feature extraction device comprising: a behavior feature extraction unit that extracts and extracts, as a workplace residence point, the residence point having the second most residence days among the plurality of residence points.
The behavior feature extracting means determines the number of stay days based on the presence / absence of the stay every 24 hours based on the stay start date and time of the first stay in one day in each of the plurality of stay points. The behavior feature extraction apparatus according to claim 1, wherein
The behavior feature extraction means calculates a total residence time for each of the stay points having the most staying days when there are a plurality of stay points having the most stay days among the plurality of stay points. The residence point with the longest total residence time is extracted as the home residence point among the residence points with the longest residence days, and the total residence time is 2 among the residence points with the largest residence days. The behavior feature extraction device according to claim 1 or 2, wherein the second longest stay point is extracted as the workplace stay point.
When the number of days included in the stay point information is equal to or less than a predetermined number of days, the behavior feature extracting unit calculates a night stay number of stays that is a stay number of stays including a predetermined time at night for each of the plurality of stay points. Then, out of the plurality of staying points, the staying point having the largest number of staying days at night is extracted as the home staying point, and among the plurality of staying points, the number of staying days is largest except for the home staying point. The behavior feature extraction apparatus according to claim 1, wherein the stay point is extracted as the workplace stay point.
The behavior feature extracting means calculates a residence time per day for each of the plurality of residence points, and the residence point having a residence time per day of a predetermined time or more among the plurality of residence points. The behavior feature extraction device according to claim 1, wherein the stay days are calculated.
The stay point information further includes a stay data identifier for each of the plurality of stays,
The behavior feature extraction means includes
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time, the residence point analysis start point information is the last of the home residence point or the workplace residence point. Update with the stay data identifier of the stay performed after the stay,
Using the stay point information after the stay indicated by the stay data identifier included in the stay point analysis start point information, the home stay point and the workplace stay point are extracted.
The behavior feature extraction device according to claim 1.
The behavior feature extraction means includes
When the stay at the home residence point or the workplace residence point extracted last time is not performed for a predetermined period until the current date and time,
If the home residence point extracted this time does not match the home residence point before update, or if the workplace residence point extracted this time does not match the workplace residence point before update, the residence point analysis start point information in the previous extraction The previously extracted home residence point extracted previously and the workplace residence point extracted last time are set as the before-update residence point analysis start point information, the home residence point before update, and the workplace residence point before update, respectively.
If the home residence point extracted this time matches the home residence point before update, and the workplace residence point extracted this time matches the workplace residence point before update, the residence point analysis start point information before the update The behavior feature extraction apparatus according to claim 6, wherein the stay point analysis start point information is set.
A terminal having position information acquisition means for acquiring a positioning point indicating the position of the user together with the positioning date and time and outputting position information including the positioning point and the positioning date and time;
Based on the position information, the stay point information including each stay point of the plurality of stays of the user, the stay start date and time at the stay point, and the stay end date and time at the stay point is extracted and stored in the behavior type storage unit. A behavior type extraction device having a behavior type extraction means;
The stay point information is acquired from the behavior type extraction device, the stay days are calculated for each of the plurality of stay points based on the stay point information, and the stay days are the largest among the plurality of stay points. A behavior feature extraction device having behavior feature extraction means for extracting the stay point as a home stay point, and extracting the stay point having the second most stay days among the plurality of stay points as a workplace stay point. Behavior feature extraction system.
Based on residence point information including each residence point of the plurality of residences of the user, residence start date and time at the residence point, and residence end date and time at the residence point, the residence days are calculated for each of the plurality of residence points, Among the plurality of staying points, the staying point having the most staying days is extracted as a home staying point, and among the plurality of staying points, the staying point having the second most staying days is extracted as a workplace staying point. Action feature extraction method.
On the computer,
Based on residence point information including each residence point of the plurality of residences of the user, residence start date and time at the residence point, and residence end date and time at the residence point, the residence days are calculated for each of the plurality of residence points, Among the plurality of staying points, the staying point having the most staying days is extracted as a home staying point, and among the plurality of staying points, the staying point having the second most staying days is extracted as a workplace staying point. The computer-readable recording medium which stores the action feature extraction program which performs the process to perform.