JP2010277190A - Device and method for extraction of staying place context, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device and method for extraction of a staying place context, and a program, for successively extracting a stay section regardless of whether the staying place is indoors or outdoors from continuously collected movement history information. <P>SOLUTION: The staying place context extraction device 100 acquires a GPS log (log information) and the access point detection history of a radio LAN device as movement history information from a location information collection device 200, and stores the movement history information in a movement history information DB141. Then, the place of context extraction device 100 acquires a GPS log from the movement history information DB141 with prescribed intervals, and when a distance and speed between two data are equal to or less than a prescribed value, extracts the staying place section by using the latitude and longitude of the acquired information, the mean value of detection dates and each mean value of the previous sections. Also, the place of context extraction device 100 acquires an access point detection history from the movement history information DB141, and extracts the section in which the same access point information continues in a prescribed time as a staying place section. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a stay location context extraction apparatus, a stay location context extraction method, and a program.

To automatically grasp human behavior history when promoting behavior recall for diary blogs, accumulating behavior for automatic creation of personal history, behavior analysis for safe and efficient behavior recommendations, etc. There are various requirements. Among them, it is possible to sequentially extract staying places that are assumed to be meaningful places without omission from the movement history data that is continuously extracted using devices that are close to each other, without any omissions. , Especially important.
The prior art relating to this requirement includes the following methods.

  For example, in Non-Patent Document 1, GPS data periodically accumulated using a GPS (Global Positioning System) logger is processed using a data categorization technique, that is, an area where data is concentrated, that is, stayed. A technique for extracting a wax location is disclosed.

  In Non-Patent Document 2, location information (latitude and longitude) is added to all access points of a wireless LAN device (WiFi (registered trademark): Wireless Fidelity) and managed in advance on the server side. A technique is disclosed in which when one or more collected WiFi data is transmitted to a server, a current position is estimated from the data using a technique such as triangulation.

Nishino, 5 others, “Examination of behavior pattern extraction method from stay transition information”, Information Processing Society of Japan, 20th UBI Study Group, November 13, 2008 Junichi Kokimoto, Takashi Ami Yagi “When becomes Where: Continuous Location History Acquisition and Its Application Using WiFi Self-Logging”, Information Processing Society of Japan Interaction 2007

  However, the technique described in Non-Patent Document 1 is not suitable for the sequential extraction of the stay place information in real time because the stay place cannot be extracted with high accuracy unless data is accumulated to some extent. Also, since linear interpolation is used for complementing the valleys of buildings where GPS data acquisition is lacking and indoor GPS data, if it is missing at the indoor entrance and GPS data is re-supplemented at the same point, the entrance becomes the place of stay There were problems such as incorrect extraction.

  Also, with the technique described in Non-Patent Document 2, the current location cannot be estimated in a method, for example, in a remote place where access points of wireless LAN devices cannot be collected. In addition, the location of the access point is not easily assumed, and there is a problem that it takes a lot of trouble to accurately register latitude and longitude information in advance.

  The present invention was made in view of such a background, and the present invention sequentially extracts stay information regardless of whether indoors or outdoors, from movement history data continuously collected using nearby devices. An object of the present invention is to provide a place of stay context extraction device, a place of stay context extraction method, and a program.

  In order to solve the above-described problem, the invention according to claim 1 is connected to a position information collecting apparatus including a satellite radio wave communication unit and a wireless network communication unit via a communication network, and receives radio waves from the satellite radio wave communication unit. Log information consisting of the user's latitude, longitude and the detected date and time of detecting the latitude and longitude of the user of the position information collecting device sequentially detected based on the information, and the wireless communication network detected by the wireless network communication means The access point information and the access point detection history including the detection date and time of the access point information are received from the location information collection device as the user's movement history information, and the user stays using the received movement history information. Stay indicating the stay section during the user's movement based on the stay location information and stay time It is a stay location context extraction device that generates interval information, a movement history information database in which the movement history information is stored, a storage section that stores a stay section information database in which the stay section information is stored, and the position information The travel history information is received from the collection device and stored in the travel history information database, and the stay section information is generated and stored in the stay section information database using the plurality of travel history information. A stay section extracting unit that acquires the plurality of log information of the user from the travel history information stored in the travel history information database, and the plurality of acquired logs The information is divided into predetermined time intervals, and the log information group belonging to the divided section is sorted in the order of the oldest section. In order to determine whether or not to include in the section, the average value of the latitude and longitude of the log information group and the detection date and time of the log information is calculated, and the calculated average value is used as the log information group of the next section. Is sequentially overwritten and stored as immediately preceding section information for determining whether to include in the stay section, and each average value of the latitude and longitude shown in the log information group in the next section and the detection date and time of the log information is calculated. Then, from each calculated average value and each average value stored as the immediately preceding section information, the distance and speed between the section and the immediately preceding section are calculated, and the calculated distance is less than a predetermined value, When the calculated speed is less than a predetermined value, the log information group of the section and the log information group of the immediately preceding section are combined to calculate the average value, and the calculated information is used as the first stay section information. Extract and further The plurality of access point detection histories of the user are acquired from the movement history information stored in the movement history information database, and the access point detection history group in which the same access point information continues for a predetermined time is stored in the second stay. The stay section information is generated by adding the second stay section information of the time that cannot be acquired as the stay section information extracted as section information and the time that cannot be acquired as the first stay section information. It is characterized by that.

  The invention according to claim 4 is connected to a position information collecting device including a satellite radio wave communication unit and a radio network communication unit via a communication network, and is sequentially detected based on radio wave information from the satellite radio wave communication unit. Log information including the latitude and longitude of the user of the position information collection device and the detection date and time when the latitude and longitude are detected, access point information of the wireless communication network detected by the wireless network communication means, and access point information thereof The access point detection history consisting of the detection date and time is received from the location information collection device as the user's movement history information, and using the received movement history information, the stay location information and the stay time where the user stayed The stay destination controller that generates stay section information indicating the stay section during the movement of the user A stay location context extracting method used in a kist extracting device, wherein the staying device context extracting device includes a travel history information database in which the travel history information is stored, and a stay section information database in which the stay section information is stored. A storage unit for storing, receiving the movement history information from the position information collection device, and storing the movement history information in the movement history information database; and generating the stay section information using the plurality of movement history information; Storing in the stay section information database, and further generating the stay section information from the travel history information stored in the travel history information database. And dividing the acquired plurality of log information into predetermined time intervals, In order to determine whether or not to include in the stay section, the log information group belonging to the section in the order of the oldest time, the average value of the latitude and longitude of the log information group and the detection date and time of the log information is calculated. Each of the calculated average values is sequentially overwritten and stored as immediately preceding section information for determining whether to include the log information group of the next section in the stay section, and in the log information group of the next section Calculate each average value of the indicated latitude, longitude, and the detection date and time of the log information, and calculate the average value between the calculated average value and each average value stored as the immediately preceding section information. When the calculated distance is less than a predetermined value and the calculated speed is less than a predetermined value, the log information group of the section and the log information group of the immediately preceding section are combined to A step of calculating each average value, extracting the calculated information as first stay section information, and detecting the plurality of access points of the user from the movement history information stored in the movement history information database A step of obtaining a history, extracting an access point detection history group in which the same access point information continues for a predetermined time as second stay section information, and stay section information at a time when the first stay section information could not be obtained. And generating the stay section information by adding the second stay section information of the time that could not be acquired.

By doing in this way, the place of stay context extraction device can receive log information and access point detection history from the location information collection device as movement history information. And the stay place context extraction device divides the plurality of log information acquired from the received movement history information at predetermined time intervals, and the average value of the latitude, longitude, and detection date and time of the log information group for each divided section Calculate Then, the distance and speed between the section and the immediately preceding section are calculated from the calculated average value and the average value of the latitude, longitude, and detection date / time of the log information group calculated in the immediately preceding section. Subsequently, when the calculated distance is less than the predetermined value and the calculated speed is less than the predetermined value, the stay location context extracting device combines the log information group of the section and the log information group of the immediately preceding section. An average value is calculated, and the calculated information is extracted as first stay section information.
Further, the stay location context extraction apparatus acquires an access point detection history from the received movement history information, and extracts an access point detection history group in which the same access point information continues for a predetermined time as a second stay section.
And a stay location context extraction apparatus adds 2nd stay area information as stay area information in the time which was not able to be acquired as a 1st stay area, and produces | generates stay area information.

  That is, the stay location context extraction device acquires the movement history information from the position information collection device provided with the satellite radio wave communication means and the wireless network communication means, and uses the movement history information obtained from the log information outdoors, Using the travel history information obtained from the access point detection history, it is possible to sequentially extract stay sections.

  According to a second aspect of the present invention, in the stay location context extracting apparatus according to the first aspect, two pieces of stay section information adjacent to each other in the average value of the calculated log information detection dates and times are extracted from the stay section information database. The interval between the average values of the detection times of the log information of the extracted two stay section information is equal to or less than a predetermined value, and is indicated by the latitude and longitude of the log information group of each of the two extracted stay section information A stay section connection unit is provided that stores the two stay section information as one stay section information in the stay section information database when the stay destination areas overlap.

  By doing in this way, the stay area information extracted by the stay place context extraction device can be corrected. Specifically, the stay location context extraction device can link two pieces of stay section information having a close detection date and a close position as one piece of stay section information. Therefore, it is possible to extract two adjacent stay section information as one stay section.

  According to a third aspect of the present invention, in the stay location context extracting device according to the second aspect, the stay section information is extracted from the stay section information database, and a length from a start time to an end time of the stay section is determined. A stay section deleting unit that deletes the stay section information from the stay section information database when the stay section information is less than a predetermined value is provided.

  By doing in this way, even if it is information extracted as stay section information, it is possible to delete stay section information determined that the stay time is short and need not be recorded as stay information. it can.

  According to a fifth aspect of the present invention, there is provided a program for causing a computer to execute the stay destination context extracting method according to the fourth aspect.

  According to such a program, the place of stay context extraction method according to claim 4 can be executed by a general computer.

  According to the present invention, a stay location context extraction apparatus, a stay location context extraction method, and a program that sequentially extract stay information regardless of whether indoors or outdoors, from movement history data continuously collected using a nearby device Can be provided.

It is a figure which shows the whole structure of the stay place context extraction system which concerns on this embodiment. It is a conceptual diagram explaining the extraction method of the stay area by the stay place context extraction system which concerns on this embodiment. It is a functional block diagram which shows the structural example of the stay place context extraction system which concerns on this embodiment. It is a flowchart which shows the whole flow of the extraction process of the stay area by the stay place context extraction apparatus which concerns on this embodiment. It is a flowchart which shows the flow of the extraction process of the stay area information which concerns on this embodiment. It is a flowchart which shows the flow of the stay area determination process which concerns on this embodiment. It is a flowchart which shows the flow of the correction process of the stay area information which concerns on this embodiment. It is a flowchart which shows the flow of the connection process of the stay area information which concerns on this embodiment. It is a flowchart which shows the flow of the WiFi stay area information extraction process which concerns on this embodiment. It is a flowchart which shows the flow of the deletion process of the stay area information which concerns on this embodiment. It is a flowchart which shows the flow of the addition process of the stay location information which concerns on this embodiment. It is a figure which shows an example of the movement history information which concerns on this embodiment. It is a figure which shows an example of the movement history information which concerns on this embodiment. It is a figure which shows an example of the immediately preceding area information which concerns on this embodiment. It is a figure which shows an example of the stay area information which concerns on this embodiment. It is a figure which shows an example of the area information which concerns on this embodiment. It is a figure which shows an example of the user information which concerns on this embodiment. It is a figure which shows an example of the stay area information which concerns on this embodiment. It is a figure which shows an example of the immediately preceding area information which concerns on this embodiment. It is a figure which shows an example of the stay area information list which concerns on this embodiment. It is a figure which shows an example of the stay area information list which concerns on this embodiment. It is a figure which shows an example of the stay area information list which concerns on this embodiment. It is a figure which shows an example of the stay area information which concerns on this embodiment. It is a figure which shows an example of the area information which concerns on this embodiment. It is a figure which shows an example of the area-staying section information which concerns on this embodiment.

  Next, modes for carrying out the present invention (referred to as “embodiments”) will be described in detail with reference to the drawings as appropriate.

FIG. 1 is a diagram showing an overall configuration of a stay destination context extraction system 1 according to the present embodiment.
As shown in FIG. 1, in the stay location context extraction system 1 according to the present embodiment, a stay location context extraction device 100 is connected to a communication network 400 such as the Internet, and the communication network 400 collects user location information. Devices 200 (200a, 200b, 200c,...) Are connected.

  Here, the position information collection device 200 includes a GPS (satellite radio wave communication means) or a wireless LAN communication means (wireless network communication means), and transmits the collected information to the stay location context extraction device 100. . The position information collection device 200 is, for example, a device in which the GPS logger 201 is connected to a portable terminal equipped with both GPS and wireless LAN communication means, a PC (Personal Computer) equipped with wireless LAN communication means, or the like. .

  The landmark search device 300 connected to the stay location context extraction device 100 via the communication network 400 is used, for example, as an external cloud service. When the center (latitude, longitude) and the search radius are passed, Provides landmark information search results up to the search radius in order of proximity to the specified center. The stay location context extraction apparatus 100 is connected when it is desired to add landmark information to the extracted stay information.

  Although not shown in FIG. 1, a server that uses the extracted stay information to provide a service is connected to the communication network 400, acquires the extracted stay information from the stay place context extraction device 100, and the user To refer to. This service includes support for creating diary blogs, support for creating personal history, and recommending actions.

FIG. 2 is a conceptual diagram for explaining a method of extracting a stay section by the stay place context extraction system 1 according to the present embodiment.
As shown in FIG. 2, according to the stay location context extraction system 1 according to the present embodiment, when the user is outdoors, the stay information is extracted based on information from the GPS logger included in the position information collection device 200. In addition, when there is a user indoors where GPS cannot be used, stay information is extracted using wireless LAN communication means.

  In addition, the stay location context extraction apparatus 100 according to the present embodiment extracts stay information with a short movement distance and a low movement speed as a stay section based on the movement history from the position information collection apparatus 200. And, for example, stay places such as A station, B station, workplace, convenience store (convenience store), and C park can be extracted as stay section information indicated by latitude, longitude, and radius.

FIG. 3 is a functional block diagram illustrating a configuration example of the stay destination context extraction system 1 according to the present embodiment.
As illustrated in FIG. 3, the stay location context extraction apparatus 100 according to the present embodiment includes position information (movement history information) collected by the GPS log collection unit 220 and the WiFi access point detection log collection unit 230 of the position information collection device 200. Is received via the communication network 400 by the processing of the position information transmission unit 210. In addition, the stay location context extraction apparatus 100 accesses the landmark search apparatus 300 via the communication network 400 when the landmark information of the stay section is required.

The stay location context extraction device 100 obtains the stay section information, which is a general stay location, from the GPS log (log information) collected from the location information collection device 200 provided by the user and the access point detection history of WiFi. Is extracted as
As shown in FIG. 3, the stay location context extracting apparatus 100 includes a control unit 10, a communication unit 11, an input / output unit 12, a memory unit 13, and a storage unit 14.

  The control unit 10 acquires the movement history information received from the position information collection device 200, performs a process of extracting a stay section, and has a position information acquisition unit 110 and a stay section extraction unit 120. The function of the control unit 10 is realized by, for example, a CPU (Central Processing Unit) developing and executing a program stored in the storage unit 14 of the stay location context extracting apparatus 100 in the memory unit 13.

  The location information acquisition unit 110 receives movement history information (see FIGS. 12 and 13) from the location information collection device 200 via the communication unit 11, and is stored in a user information database (DB: DataBase) 147 described later. Based on the user information (see FIG. 17), after performing user authentication, the movement history information that has been successfully authenticated is stored in the storage unit.

  The stay section extraction unit 120 extracts stay section information (see FIG. 15) from the movement history information, and performs a process of correcting the extracted stay section information. The stay section extraction unit 120 includes a stay section determination unit 121, a stay section connection unit 122, a WiFi history processing unit 123, a stay section deletion unit 124, and a stay place information addition unit 125. .

  The stay section determination unit 121 determines any one of “start of stay section”, “continuation of stay section”, and “end of stay section extraction processing” because it is not a stay section for the GPS data extracted from the movement history information. . When the stay section determination unit 121 determines that “the start of the stay section”, the stay section ID is acquired and the stay section information is generated.

  The stay section connecting unit 122 extracts two stay section information as one piece of stay section information when the time interval of the detection date and time is within a predetermined value and the distance is within a predetermined value for the extracted two stay section information. Connect as

  The WiFi history processing unit 123 acquires a WiFi access point detection history and performs processing to generate stay section information.

  The stay section deleting unit 124 calculates a stay time for the extracted stay section information, and performs a process of deleting the stay section information when the stay time is less than a predetermined value.

  The stay location information adding unit 125 searches the landmark search device 300 (see FIG. 3), the area information DB 144 and the address DB 145 described later, acquires the area information ID, and associates the area information ID with the stay section ID. -Generate stay section information (see FIG. 25).

Next, the communication unit 11 includes an interface for transmitting / receiving information to / from each device via the communication network 400.
The input / output unit 12 includes a keyboard and the like, and includes an input unit that gives an instruction regarding processing of the control unit 10 and an output unit that includes a liquid crystal display.
The memory unit 13 includes storage means such as a RAM (Random Access Memory), and temporarily stores information necessary for the control of the control unit 10.

  The storage unit 14 includes a movement history information DB 141 that stores movement history information (see FIGS. 12 and 13) that is information obtained from a GPS log (log information) and a WiFi access point detection history, and a section processed immediately before. Immediately preceding section information DB 142 storing immediately preceding section information (see FIG. 14), stay section information DB 143 storing stay section information (see FIG. 15) relating to the extracted stay section, and staying The area information DB 144 for storing area information for managing the name of the place (see FIG. 16), the address DB 145 for storing correspondence information between latitude, longitude and address, and the relationship between the stay section and the area information are managed. Area-stay section DB 146 in which area-stay section information (see FIG. 25) is stored, and user information DB1 in which user information (see FIG. 17) is stored. Configured to include a 7. The specific data structure of these pieces of information will be specifically described later.

  Next, the place of stay context extraction method by the place of stay context extraction apparatus 100 according to the present embodiment will be specifically described.

FIG. 4 is a flowchart showing the overall flow of the stay section extraction process by the stay place context extraction apparatus 100 according to the present embodiment.
As shown in FIG. 4, first, the location information acquisition unit 110 of the stay location context extraction device 100 receives user information and a GPS log (log information) or access point detection from the location information collection device 200 via the communication unit 11. The history is received as movement history information (step S1000).

Here, the user information (see FIG. 17) is information composed of a user name and a password.
Further, the movement history information (see FIGS. 12 and 13) is acquired from the user ID that is a unique number of the user, the type indicating whether the received information is GPS data or WiFi data, the detection date and time, and the GPS. Latitude, longitude, MAC address acquired from the access point detection history of WiFi, radio wave intensity, and SSID (Service Set Identifier) that is an identifier of a wireless LAN access point.

Next, the position information acquisition unit 110 authenticates user information for determining whether or not the acquired user name and password are registered in the user information DB 147 (step S2000).
If the user information is registered in the user information DB 147 and the authentication is successful (step S3000 → Yes), the process proceeds to the next step S4000. On the other hand, if the authentication fails (step S3000 → No), the stay section information extraction process ends as an error (step S6000).

  In step S4000, the position information acquisition unit 110 registers the movement history information (see FIGS. 12 and 13) that has been successfully authenticated by the movement history information DB 141 in the storage unit 14.

  Then, the stay section extracting unit 120 performs a process of extracting a stay section (stay section information extracting process) using the travel history information registered in the travel history information DB 141 (step S5000). The stay section information extraction process will be described in detail with reference to FIG.

(Extraction processing of stay section information)
FIG. 5 is a flowchart showing the flow of the stay segment information extraction process according to the present embodiment.
First, the stay section extraction unit 120 acquires immediately preceding section information, which is information of the immediately preceding stay section, from the immediately preceding section information DB 142 using the user ID as a key (step S5100).

The immediately preceding section information (see FIG. 14) includes a user ID, an arithmetic average of latitude and longitude of data detected in the immediately preceding section, an arithmetic average of detection date and time, a stay section flag, and the data It has a start date and time, an end date and time, and a stay section ID.
Here, the stay section flag is information indicating whether or not the immediately preceding section is extracted as the stay section. If the immediately preceding section is extracted as the stay section, “true” is input. "False" is entered. In addition, when the stay section flag is “true”, the stay section ID attached to the stay section is input, and when it is “false”, “null” is input.

Next, the stay section extraction unit 120 extracts a GPS log from the movement history information DB 141 (step S5200).
And the stay section extraction part 120 acquires the GPS log for every predetermined | prescribed time interval (section) in order of data oldness (step S5300). The predetermined time interval (section) is set in advance, for example, as a 1-minute interval, a 3-minute interval, or the like.

  Subsequently, the stay section extraction unit 120 determines whether or not the number of acquired data is 0 (step S5400). If the acquired number is not 0 (step S5400 → No), the process proceeds to step S5500.

  In step S5500, stay section determination unit 121 determines whether the acquired data corresponds to the start of a new stay section, the continuation of the stay section, or the end of the stay section extraction process (stay section determination process). The stay section determination process will be described in detail with reference to FIG.

  Next, when the number of acquired data is 0 in step S5400, or when the stay section determination process in step S5500 is finished, the stay section extraction unit 120 determines whether or not all GPS logs have been processed. Judgment is made (step S5600). If there is a GPS log that has not yet been processed (step S5600 → No), the process returns to step S5300 and continues. On the other hand, when all the GPS logs have been processed (step S5600 → Yes), the process proceeds to the next step S5700 to perform stay section information correction processing (see FIG. 7 described later).

(Stay section judgment processing)
Next, the stay section determination process performed by the stay section determination unit 121 in step S5500 in FIG. 5 will be described.
FIG. 6 is a flowchart showing the flow of stay period determination processing according to the present embodiment.
First, the stay section determination unit 121 calculates an arithmetic average of latitudes, an arithmetic average of longitudes, and an arithmetic average of detection dates and times acquired within a predetermined interval in step S5300 of FIG. 5 (step S5510).

  Subsequently, the stay section determination unit 121 calculates the average value calculated in step S5510, the arithmetic average value of the latitude of the immediately preceding section information (see FIG. 14), the arithmetic average value of the longitude, and the arithmetic average value of the detection date and time. From the above, the distance and speed moved are calculated (step S5520). The immediately preceding section information is stored by overwriting the immediately preceding section information DB with information on the section acquired and processed immediately before the section acquired at a predetermined time interval in step S5300 in FIG. (See steps S5540, S5570, and S5590 described later).

Then, stay section determination unit 121 determines whether or not the calculated distance is less than a predetermined value and the speed is less than a predetermined value (step S5530).
If the distance is less than the predetermined value and the speed is not less than the predetermined value (step S5530 → No), the stay section extraction process for the GPS log is terminated (end of the stay section extraction process). At that time, the data calculated in step S5510 is stored in the immediately preceding section information, the stay section flag is set to “false”, the stay section ID is set to “null”, and the immediately preceding section information DB 142 is overwritten (step S5540).

  On the other hand, if the distance is less than the predetermined value and the speed is less than the predetermined value (step S5530 → Yes), the process proceeds to step S5550 to generate or update the stay section information of the acquired data.

In step S5550, when the stay section flag of the immediately preceding section information is “false” (step S5550 → Yes), the stay section information is generated using the data calculated in step S5510 and the immediately preceding section information. The ID is acquired and stored in the stay section information DB 143 (step S5560).
Then, stay section determination unit 121 overwrites the previous section information DB with the data calculated in step S5510 (step S5570). At that time, the stay section flag is set to “true” and stored.
The stay section information generated based on the GPS log data is referred to as first stay section information in the claims.

In step S5550, when the stay section flag of the immediately preceding section information is not “false” (that is, “true”) (step S5550 → No), the stay section information is obtained using the data calculated in step S5510 and the immediately preceding section data. It is updated and stored in the stay section information DB 143 (step S5580).
Then, stay section determination unit 110 overwrites the previous section information DB with the data calculated in step S5510 (step S5590). At that time, the stay section flag is set to “true” and stored.

This stay section information (see FIG. 15) includes a stay section ID that is a unique number assigned to each stay section, a user ID, a type indicating whether the received information is GPS data or WiFi data, and latitude And an arithmetic average value of longitude, a radius, an arithmetic average value of detection date / time, a WiFi area name, and a start date / time and end date / time of acquired data.
Here, the radius is calculated by calculating the center of gravity from the position information of the GPS tag data group of the date and time in the stay section and wrapping all the data. When the type is WiFi data, the latitude, longitude, and radius values are “null”, and the WiFi area name is input. In the case of stay section information using GPS data, the WiFi area name is “null”.

(Correction processing for stay section information)
Next, the stay section information correction process in step S5700 of FIG. 5 will be described.
FIG. 7 is a flowchart showing the flow of the correction process for the stay section information according to the present embodiment.
First, the stay section extraction unit 120 acquires stay section information from the stay section information DB 143, and generates a stay section information list (see FIG. 20) in the memory unit 13 (see FIG. 3) (step S5710).

  Next, in step S5720, the stay section connecting unit 122 reads two pieces of stay section information that are consecutive in chronological order, and connects the two stay section information based on the detection date and time interval and the distance between the centroids of the stay sections. A process (connection process of stay section information) is performed (see FIG. 8 described later).

  In step S5730, the WiFi history processing unit 123 acquires a WiFi access point detection history (WiFi data), and performs stay section information extraction processing (see FIG. 9 described later).

  Subsequently, in step S5740, the stay section deletion unit 124 calculates the length (time) of the stay section from the stay section information, and deletes the stay section information when the time is less than a predetermined value (stay section information). (See FIG. 10 to be described later).

  Next, in step S5750, the stay location information adding unit 125 combines the stay section information and the area information, and performs a stay place information adding process for adding an area name to the stay section (see FIG. 11 described later).

(Consolidation processing of stay section information)
Next, the stay segment information linking process in step S5720 of FIG. 7 will be described.
FIG. 8 is a flowchart showing a flow of stay interval information connection processing according to the present embodiment.
First, the stay section connection unit 122 reads two stay section information in the order of old date and time in the stay section information list generated in step S5710 of FIG. 7 (step S5721). And stay section connection part 122 ends processing, if it cannot read (Step S5722-> No). On the other hand, if the reading has been completed (step S5722 → Yes), the process proceeds to the next step S5723.

  Next, the stay section connection unit 122 determines whether the time between the detection dates of the two stay section information is equal to or less than a predetermined value (step S5723). If it is not less than the predetermined value (step S5723 → No), the process returns to step S5721 and continues. If it is below the predetermined value (step S5723 → Yes), the process proceeds to step S5724.

  And the stay section connection part 122 has an area | region which overlaps with the stay place (circle area | region calculated | required by a radius using latitude and longitude as a gravity center) shown by the latitude and the longitude obtained from the GPS log of each of the read two stay section information It is determined whether or not. Specifically, it is determined whether the sum of the radii is larger than the distance between the centers of gravity of the two stay section information (step S5724). If the sum of the radii is smaller (step S5724 → No), the process returns to step S5721 and continues. On the other hand, if the sum of the radii is larger (step S5724 → Yes), it is determined that there is an area overlapping the stay section, and the process proceeds to next step S5725.

In step S5725, the stay section connection unit 122 connects the two stay section information read in step S5721, overwrites the stay section information list, and stores the information in the stay section information DB 143.
Specifically, the end date / time of the previous stay section information is overwritten with the end date / time of the stay section information, the stay section information is deleted from the list, and the GPS data of the date / time in the stay section is obtained from the movement history information. Obtaining, calculating the center of gravity, calculating a radius including the whole, and overwriting the stay section information list and the stay section information DB 143 for the previous stay section information.

(WiFi stay section information extraction process)
Next, the WiFi stay section information extraction process in step S5730 of FIG. 7 will be described.
FIG. 9 is a flowchart showing a flow of WiFi stay section information extraction processing according to the present embodiment.
First, the WiFi history processing unit 123 acquires a WiFi access point detection history (see FIGS. 12 and 13) from the movement history information DB 141 (step S5731).

  Next, the WiFi history processing unit 123 determines whether or not the access point detection history can be acquired (step S5732). If the access point detection history cannot be acquired (step S5732 → No), the process ends. On the other hand, if it can be obtained (step S5732 → Yes), the process proceeds to step S5733. At this time, the one with the same detection date and time has the maximum strength (radio wave strength).

Subsequently, in step S5733, the WiFi history processing unit 123 extracts a section in which the same access point information (MAC address) continues for a predetermined time or more as a stay section. Here, even with different MAC addresses, access point information having the same WiFi area name is extracted as the same stay section.
The stay section information generated based on the data of the access point detection history is referred to as second stay section information in the claims.

  Then, the WiFi history processing unit 123 registers the extracted stay section in the stay section information list (step S5734), and stores the extracted stay section in the stay section information DB 143 (step S5735).

  By doing in this way, for example, even when the user has moved indoors or underground and has not been able to generate stay section information (first stay section information) from the GPS log, the stay section information can be obtained from the access point detection history. (Second stay section information) can be generated and added to generate stay section information.

(Deleting section information)
Next, the stay section information deletion process in step S5740 of FIG. 7 will be described. Even if the stay segment information is deleted as the stay segment information, the stay segment information is determined to have a short stay time and is not required to be recorded as the stay information. The information is deleted from the information DB 143.

FIG. 10 is a flowchart showing the flow of the stay section information deletion process according to this embodiment.
First, the stay section deletion unit 124 reads stay section information in order from the top of the stay section information list (step S5741). Then, it is determined whether or not there is read stay section information (step S5742). If all the stay section information is processed and there is no stay section information to be read (step S5742 → No), the process ends. On the other hand, if there is read stay section information (step S5742 → Yes), the process proceeds to the next step S5743.

  In step S5743, the stay section deletion unit 124 calculates the length (time) of the stay section in the stay section information, and determines whether the length is less than a predetermined value (step S5743). And when the length of a stay area is not less than predetermined value (step S5743-> No), it returns to step S5741 and continues a process. On the other hand, when the length of the stay section is less than the predetermined value (step S5743 → Yes), the stay section information is deleted from the list and is also deleted from the stay section information DB 143 (step S5744).

(Additional processing of stay value information)
Next, the stay location information addition processing in step S5750 of FIG. 7 will be described.
FIG. 11 is a flowchart showing the flow of the stay location information adding process according to the present embodiment.
First, the stay location information adding unit 125 reads the stay section information in order from the head of the stay section information list (step S5751). Then, it is determined whether there is read stay section information (step S5752). If all the stay section information is processed and there is no stay section information to be read (step S5752 → No), the process ends. On the other hand, if there is read stay section information (step S5752 → Yes), the process proceeds to the next step S5753.

  Next, the stay location information adding unit 125 determines whether or not the type of stay section information is GPS (step S5753). If the type is GPS (step S5753 → Yes), the process proceeds to step S5754. If the type is not GPS but WiFi (step S5753 → No), the process proceeds to step S5757.

When the stay section is extracted from the GPS log (step S5753 → Yes), the stay place information adding unit 125 searches for the area name of the stay section by the following three methods.
First, the stay location information adding unit 125 searches using the area information DB 144 in the storage unit 14 using the latitude, longitude, and radius as keys, and if there is a record with overlapping areas, obtains the record with the closest center of gravity. Then, the area information ID of the information is acquired (step S5754).

  Next, the stay location information adding unit 125 transmits information on the latitude, longitude, and search radius to the landmark search device 300, and acquires the search results of the landmark information up to the search radius in the order closer to the center. The newly acquired landmark name is registered in the area information DB 144, and the area information ID is acquired (step S5755).

  Subsequently, the stay location information adding unit 125 searches for the area name of the stay section using the address DB 145 in which the mapping information of the address, the latitude, and the longitude is stored. Then, the newly acquired address is registered in the area information DB 144, and the area information ID is acquired (step S5756).

  On the other hand, if it is determined in step S5753 that the stay section information is generated from the access point detection history of WiFi and the type is not GPS (step S5753 → No), the area information is set using the WiFi area name as a key. Search the DB 144. If there is no hit record, the area information is registered in the area information DB 144, and the area information ID is acquired (step S5757).

  Then, the stay location information adding unit 125 associates the acquired area information ID with the stay section ID of the stay section information in steps S5754 to S5757, and registers them in the area-stay section DB 146 (step S5758).

(Detailed explanation of data processing)
Next, each process will be described in more detail on the assumption that movement history information is input from the location information collection device 200 to the stay location context extraction device 100.

First, as a premise of processing performed by the stay location context extraction apparatus 100, the information stored in the storage unit 14 is that the immediately preceding section information is FIG. 14, the stay section information is FIG. 15, the area information is FIG. 16, and the user information is A registered state is assumed as shown in FIG.
Then, user authentication is performed using the user information (FIG. 17) by the processing of the position information acquisition unit 110 (step S2000 in FIG. 4), and the processing of step S4000 is included in the movement history information DB 141 in the storage unit 14. Therefore, description will be made assuming that the movement history information shown in FIGS. 12 and 13 is registered. 12 and 13 are processed and registered at a time in step S4000, and some of the data are shown in FIGS.

(Extraction processing of stay section information)
The stay section extraction unit 120 acquires the immediately preceding section information from the immediately preceding section information DB 142 using the user ID as a key (step S5100 in FIG. 5). As a result, the following immediately preceding section information is acquired.

  Then, the stay section extraction unit 120 extracts a GPS log from the movement history information DB 141 (step S5200 in FIG. 5), and acquires a GPS log for each predetermined time interval in order from the oldest data (step S5300 in FIG. 5). . In the present embodiment, the predetermined time interval is 2 minutes, and a GPS log data group (log information group) is acquired for each section as follows.

(Stay section judgment processing)
Next, the stay section determination unit 121 performs a stay section determination process (step S5500 in FIG. 5). Specifically, the stay section determination unit 121 calculates the arithmetic average value of the latitude of the extracted GPS log, the arithmetic average value of the longitude, and the arithmetic average value of the detection date (step S5510 in FIG. 6). And the following results are obtained.

  Subsequently, the stay section determination unit 121 includes the arithmetic average value of the latitude, the arithmetic average value of the longitude, the arithmetic average value of the detection date and time of the previous section information which is the information of the section acquired immediately before, as described above. The distance (km) is calculated using the result of (Table 3). Further, the speed is obtained from the distance and the time difference (step S5520 in FIG. 6). And the following results are obtained.

  Then, stay section determination unit 121 determines whether or not the calculated distance is less than a predetermined value and the speed is less than a predetermined value (step S5530 in FIG. 6). In the present embodiment, the predetermined distance value is “0.1 km” and the predetermined speed value is “2 km / h”.

Therefore, the stay section determination unit 121
(1) When the speed is less than 2 km / h, the distance is less than 0.1 km, and the stay section flag of the immediately preceding section information is “false” (step S5550 → Yes in FIG. 6), the stay section is started. The process proceeds to step S5560.
(2) When the speed is less than 2 km / h, the distance is less than 0.1 km, and the stay section flag of the immediately preceding section information is “true” (step S5550 → No in FIG. 6), the stay section is continued. The process proceeds to step S5580.
(3) In cases other than the above, the process proceeds to step S5540 in FIG.

Therefore, the stay section determination unit 121 processes the data in (Table 4) as follows.
(a) [section of 2009-4-12 04:56:45 UTC to 2009-4-12 04:58:45 UTC], the process proceeds to step S5540 in FIG. End of section extraction processing).
(b) [2009-4-12 04:58:45 UTC to 2009-4-12 05:00:45 UTC]], the process proceeds to step S5540 in FIG. End of section extraction processing).
(c) [2009-4-12 05:00:45 UTC to 2009-4-12 05:02:45 UTC section], the process proceeds to step S5560 in FIG. processing).
(d) [section of 2009-4-12 05:02:45 UTC to 2009-4-12 05:04:45 UTC] proceeds to step S5580 in FIG. processing).

(End of stay segment extraction process)
At the end of the stay section extraction process in step S5540 of FIG. 6, the stay section determination unit 121 immediately calculates the arithmetic average of the detection date, the arithmetic average of latitude, and the arithmetic average of longitude calculated in (Table 3). Overwrite the section information and set the stay section flag to “false”. Then, the oldest detection date and time of the GPS log extracted in (Table 2) of step S5300 in FIG. 5 is set as the start date and time of the immediately preceding section information, and the latest detection date and time is set as the end date and time of the immediately preceding section information. The section ID is set to “null” and the immediately preceding section information DB 142 is overwritten and saved.

  As a result of the data (a), the immediately preceding section information DB 142 is overwritten with the following immediately preceding section information.

  As a result of the data of (b) above, the immediately preceding section information DB 142 is overwritten with the following immediately preceding section information.

(Stay segment start processing)
In step S5560 of FIG. 6, the stay section determination unit 121 registers the start date and time of the immediately preceding section information as the start date and time of the stay section information, and also extracted as a result of step S5300 of FIG. 5 (Table 2). The latest date and time of the data is registered as the end date and time of the stay section information. Furthermore, the GPS data of the date and time in the stay section is acquired from the movement history information DB 141, the center of gravity is calculated, the radius that wraps the whole is calculated, and the latitude, longitude, and radius of the stay section information are stored in the stay section information DB 143. sign up.

  In addition, the stay section determination unit 121 overwrites the arithmetic average of the detection date and time, the arithmetic average of the latitude, and the arithmetic average of the longitude calculated in (Table 3) on the immediately preceding section information, and sets the stay section flag to “true” " Then, the oldest detection date and time of the GPS log extracted in (Table 2) of step S5300 in FIG. 5 is set as the start date and time of the immediately preceding section information, and the latest detection date and time is set as the end date and time of the immediately preceding section information. The section ID is set, and the immediately preceding section information DB 142 is overwritten and saved.

  As a result of the above-mentioned stay section start process, the stay section ID “100” in the section (c) of [2009-4-12 05:00:45 UTC to 2009-4-12 05:02:45 UTC] And the stay section information is registered as shown in (Table 7). Further, the immediately preceding section information is overwritten and saved as shown in (Table 8).

(Staying section continuation processing)
In step S5580 in FIG. 6, the stay section determination unit 121 uses the stay section ID of the immediately preceding section information as a key to indicate the end date of the stay section information in the stay section information DB 143 as a result of step S5300 in FIG. 5 (Table 2). Overwrite with the latest date and time of the GPS log extracted in. Furthermore, the GPS data of the date and time in the stay section is acquired from the movement history information DB 141, the center of gravity is calculated, the radius that wraps the whole is calculated, and the latitude, longitude, and radius of the stay section information are stored in the stay section information DB 143. sign up.

  In addition, the stay section determination unit 121 overwrites the arithmetic average of the detection date and time, the arithmetic average of the latitude, and the arithmetic average of the longitude calculated in (Table 3) on the immediately preceding section information, and sets the stay section flag to “true” " Then, the oldest detection date / time of the GPS log extracted in (Table 2) in step S5300 of FIG. 5 is the start date / time of the previous section information, and the latest detection date / time of the GPS log extracted in (Table 2) is the previous section information. And the stay section ID is set as it is, and the immediately preceding section information DB 142 is overwritten and saved.

  As a result of the above stay section continuation processing, the stay section ID “100” in the section (d) of [2009-4-12 05:02:45 UTC to 2009-4-12 05:04:45 UTC] The following stay section information is registered as shown in (Table 9). Further, the immediately preceding section information is overwritten and saved as shown in (Table 10).

(Correction processing for stay section information)
Next, the stay section information correction process shown in FIG. 7 will be described in detail.
First, as a premise of the stay section information correction process performed by the stay place context extraction apparatus 100, the stay section information shown in FIG. 18 and the immediately preceding section information shown in FIG. 19 are stored in the storage unit 14 in step S5200 of FIG. A registered state is obtained by extracting a GPS log from the history information DB 141.

  In step S5710 of FIG. 7, for the information of the user ID “1”, the stay section information shown in FIG. 18 stored in the storage unit 14 and the one stay section information immediately before that are acquired from the stay section information DB 143. The stay section information list shown in FIG. 20 is generated. However, in the present embodiment, the immediately preceding stay section information is zero.

(Consolidation processing of stay section information)
Next, the stay section connection unit 122 performs the stay section information connection process in step S5720 of FIG.
First, in the stay section information list (see FIG. 20) acquired in step S5710 in FIG. 7, two stay section information is read in order from the oldest date and time (step S5721 in FIG. 8).

  And the stay section connection part 122 determines whether the time between the detection dates of the stay section of the acquired two stay section information is below a predetermined value (step S5723 of FIG. 8). In this embodiment, if the predetermined period is 2 minutes, there is 20 minutes between the stay section ID “100” and the stay section ID “101”, which is larger than the predetermined value, so nothing is done. Also, between the stay section ID “101” and the stay section ID “102” is 21 minutes, which is larger than the predetermined value, so nothing is done.

(WiFi stay section information extraction process)
Next, the WiFi history processing unit 123 performs a process of extracting stay section information from the WiFi access detection history (step S5730 in FIG. 7).

  First, the WiFi history processing unit 123 acquires the access point detection history of WiFi from the movement history information DB 141 (see FIGS. 12 and 13) (step S5731 in FIG. 9). Here, the thing of the intensity | strength with the same detection date is acquired.

Next, the WiFi history processing unit 123 extracts a MAC address from the acquired access point detection history. As MAC addresses, “aa: aa: aa: aa: aa: aa”, “bb: bb
: bb: bb: bb: bb "," cc: cc: cc: cc: cc: cc: "," dd: dd: dd: dd: dd: dd "," ee: ee: ee: ee: ee:
get ee ".

  Subsequently, the WiFi history processing unit 123 acquires an area name using the area information DB 144 (see FIG. 16) stored in the storage unit 14. If there is no hit, the SSID is used as the WiFi area name. As a result, the WiFi history processing unit 123 obtains the following list.

  Then, the WiFi history processing unit 123 extracts a section in which the same access point information (MAC address) continues for a predetermined time or more as a stay section (step S5733 in FIG. 9). And the following results are obtained. Note that, even with different MAC addresses, access point information having the same WiFi area name is extracted as the same stay section.

  Subsequently, the WiFi history processing unit 123 registers the stay section extracted as described above in the stay section information list (step S5734 in FIG. 9), and further stores it in the stay section information DB 143 (step S5735 in FIG. 9). . FIG. 21 shows the result of registering the stay section extracted by the WiFi history processing unit 123 in the stay section list.

(Deleting section information)
Next, the stay section deletion unit 124 performs a process of deleting stay section information having a short stay section length (step S5740 in FIG. 7).

First, the stay section deletion unit 124 reads stay section information in order from the head of the stay section information list shown in FIG. 21 (step S5741 in FIG. 10).
Then, the stay section deleting unit 124 calculates the length of the stay section (end date-start date), and if the length is less than a predetermined value (5 minutes in the present embodiment) (step in FIG. 10). S5743-> Yes), the stay section information is deleted from the stay section information list and stay section information DB 143 (step S5744 in FIG. 10).

In the stay section information list of FIG. 21, the stay section ID “102” has a stay section length of 3 minutes and 30 seconds, which is less than a predetermined time of 5 minutes. Therefore, from the stay section information list and stay section information DB 143 delete.
Also, the stay section ID “103” is deleted because the length of the stay section is 4 minutes 30 seconds, which is less than the predetermined time of 5 minutes.
As a result, the stay section information list is as shown in FIG.

(Additional processing of staying location information)
Next, the stay location information adding unit 125 acquires the area name of the stay section, and performs the stay place information adding process for associating the stay section information with the area information (step S5750 in FIG. 7).

  First, the stay location information adding unit 125 reads the stay section information sequentially from the head of the stay section information list (step S5751 in FIG. 11). Subsequently, the stay location information adding unit 125 determines whether or not the type of stay section information is GPS (step S5753 in FIG. 11).

  If the stay location information is GPS data (step S5753 → Yes in FIG. 11), the stay location information adding unit 125 first searches the area information DB 144 using the latitude, longitude, and radius as keys (FIG. 11). Step S5754). Here, nothing is done for the stay section IDs “100” and “101” because there are no records with overlapping areas.

Next, the stay location information adding unit 125 searches for the area name of the stay section using the landmark search device 300 connected via the communication network 400 (step S5755 in FIG. 11).
Then, the following data is acquired for the stay section ID “100”.

This data is registered in the area information DB 144, and the acquired area information ID “3” is registered in the area-stay section DB 146 (step S5758 in FIG. 11).
Further, the following data is acquired for the stay section ID “101”.

  This data is registered in the area information DB 144, and the acquired area information ID “4” is registered in the area-stay section DB 146 (step S5758 in FIG. 11).

  Next, the stay location information adding unit 125 searches for the area name of the stay section using the address DB 145 in which the mapping information of the address, the latitude, and the longitude is stored (step S5756 in FIG. 7).

  On the other hand, for the stay section ID “104” whose latitude and longitude are generated from the access point detection history whose stay section is WiFi, the area name is searched from the area information DB 144 (step S5757 in FIG. 11). . As a result, the area information ID “2” is obtained for the stay section ID “104”. Then, the acquired area information ID “2” is registered in the area-staying section DB 146 (step S5758 in FIG. 11).

  As a result, the stay section information shown in FIG. 23, the area information shown in FIG. 24, and the area-stay section information shown in FIG. 25 can be obtained.

According to the stay location context extracting apparatus, the stay location context extracting method and the program according to the present embodiment, since it is intended to use a familiar device, the GPS information may be prepared by using a GPS function mobile phone or GPS logger. For the access point history information, a game machine, a PDA, a notebook PC or the like may be prepared, and the user can easily prepare an environment for implementing the present invention.
In addition, since the stay section is automatically extracted from the collected GPS data and WiFi access point information, the stay section can be extracted without omission of the user's hand.
Furthermore, it is possible to sequentially extract stay sections regardless of whether indoors or outdoors, using an algorithm capable of processing that combines GPS that is good outdoors and WiFi that is good indoors.

DESCRIPTION OF SYMBOLS 1 Stay place context extraction system 10 Control part 11 Communication part 12 Input / output part 13 Memory part 14 Storage part 100 Stay place context extraction apparatus 110 Location information acquisition part 120 Stay section extraction part 121 Stay section judgment part 122 Stay section connection part 123 WiFi History processing section 124 Stay section deletion section 125 Stay place information addition section 141 Movement history information DB
142 Immediately preceding section information DB
143 Stay section information DB
144 Area information DB
145 Address DB
146 Area-Stay Section DB
147 User information DB
200 position information collection device 201 GPS logger 300 landmark search device 400 communication network

Claims (5)

The user of the position information collecting apparatus connected via a communication network to a position information collecting apparatus provided with a satellite radio wave communication means and a wireless network communication means and sequentially detected based on radio wave information from the satellite radio wave communication means Access point detection history consisting of log information consisting of the latitude and longitude and the detection date and time when the latitude and longitude are detected, the access point information of the wireless communication network detected by the wireless network communication means and the detection date and time of the access point information And the user's travel history information from the location information collection device, and using the received travel history information, the stay section during the user's movement based on the stay location information and the stay time where the user stayed A stay location context extracting device for generating the indicated stay section information,
A storage unit that stores a travel history information database in which the travel history information is stored, and a stay section information database in which the stay section information is stored;
A position information acquisition unit that receives the movement history information from the position information collection device and stores the movement history information in the movement history information database;
A stay section extraction unit that generates the stay section information using the plurality of movement history information and stores the stay section information in the stay section information database;
The stay section extraction unit
The plurality of log information of the user is acquired from the movement history information stored in the movement history information database, the plurality of acquired log information is divided into predetermined time intervals, and belong to the divided section In order to determine whether or not to include log information in the stay section in order of oldest section, calculate the average value of the latitude and longitude of the log information group and the detection date and time of the log information, and calculate Each average value is sequentially overwritten and stored as immediately preceding section information for determining whether to include the log information group of the next section in the stay section,
Calculate each average value of the latitude and longitude shown in the log information group in the next section, and the detection date and time of the log information, from each calculated average value and each average value stored as the immediately preceding section information , Calculating the distance and speed between the section and the immediately preceding section, and when the calculated distance is less than a predetermined value and the calculated speed is less than the predetermined value, The average value is calculated by combining the log information group of the section, and the calculated information is extracted as the first stay section information,
Further, the plurality of access point detection histories of the user are acquired from the movement history information stored in the movement history information database, and an access point detection history group in which the same access point information continues for a predetermined time is set as a second Extract as stay section information,
The stay section information is generated by adding the second stay section information of the time that could not be acquired as the stay section information in the time that could not be acquired as the first stay section information. Ground context extraction device. From the stay section information database, two stay section information adjacent to the calculated average value of the detection times of the log information is extracted, and the interval between the average values of the log information detection dates of the two stay section information extracted Is equal to or less than a predetermined value, and the stay area information indicated by the latitude and longitude of the log information group of each of the extracted two stay section information overlaps the two stay section information as one stay section information. The stay area context extracting device according to claim 1, further comprising: a stay section connecting unit that is stored in the stay section information database. The stay section information is extracted from the stay section information database when the stay section information is extracted from the stay section information database and the length from the start time to the end time of the stay section is less than a predetermined value. The stay location context extracting apparatus according to claim 2, further comprising a deletion unit. The user of the position information collecting apparatus connected via a communication network to a position information collecting apparatus provided with a satellite radio wave communication means and a wireless network communication means and sequentially detected based on radio wave information from the satellite radio wave communication means Access point detection history consisting of log information consisting of the latitude and longitude and the detection date and time when the latitude and longitude are detected, the access point information of the wireless communication network detected by the wireless network communication means and the detection date and time of the access point information And the user's travel history information from the location information collection device, and using the received travel history information, the stay section during the user's movement based on the stay location information and the stay time where the user stayed Used in a stay location context extraction device that generates the indicated stay section information A that stay destination context extraction method,
The stayer context extraction device comprises:
A storage unit for storing a movement history information database in which the movement history information is stored, and a stay section information database in which the stay section information is stored;
Receiving the movement history information from the position information collection device and storing it in the movement history information database;
Generating the stay section information using the plurality of movement history information and storing the stay section information in the stay section information database; and
Furthermore, the step of generating the stay section information includes:
The plurality of log information of the user is acquired from the movement history information stored in the movement history information database, the plurality of acquired log information is divided into predetermined time intervals, and belong to the divided section In order to determine whether or not to include log information in the stay section in order of oldest section, calculate the average value of the latitude and longitude of the log information group and the detection date and time of the log information, and calculate Each of the average values is sequentially overwritten and stored as immediately preceding section information for determining whether to include the log information group of the next section in the stay section;
Calculate each average value of the latitude and longitude shown in the log information group in the next section, and the detection date and time of the log information, from each calculated average value and each average value stored as the immediately preceding section information , Calculating the distance and speed between the section and the immediately preceding section, and when the calculated distance is less than a predetermined value and the calculated speed is less than the predetermined value, Calculating each average value by combining log information groups of sections, and extracting the calculated information as first stay section information;
Further, the plurality of access point detection histories of the user are acquired from the movement history information stored in the movement history information database, and an access point detection history group in which the same access point information continues for a predetermined time is set as a second Extracting as stay section information;
Executing the step of generating the stay section information by adding the second stay section information of the time that could not be acquired as the stay section information in the time that could not be acquired as the first stay section information. The stay place context extraction method characterized by this. The program for making a computer perform the place of stay context extraction method of Claim 4.

Images