JP5536485B2 - Portable terminal, server, program, and method for estimating address / location as user moves - Google Patents

Description

  The present invention relates to a technique for acquiring a position accompanying movement of a user using a positioning function mounted on a mobile terminal.

  In recent years, a mobile terminal typified by a mobile phone has generally been equipped with a positioning function such as GPS (Global Positioning System). Therefore, the user can measure the current position using the mobile terminal, and can receive various service information corresponding to the position by transmitting the position to the server via the network.

  Conventionally, there is a technique in which a positioning position of a mobile terminal is transmitted to a server, and the server detects an area where a user who owns the mobile terminal frequently stays (see, for example, Patent Document 1). According to this technique, in order to detect the stay frequency for each area, it is necessary to divide the area into area units in advance. For example, it is necessary to divide an area like Tokyo into a certain mesh. This technology is used for different purposes and effects depending on how the areas are divided.

JP 2003-44512 A

Toyama Midori, Hattori Takashi, Kanno Tatsuya, "Learning Significant Positions Using Cellular Phone Positioning Functions", IPSJ Journal 46 (12), pp.2915-2924, 2005 Ester M., et.al: “A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise”, Proc. 2nd int. Conf. On Knowledge Discovery and Data Mining (KDD'96), Portland, Oregon, 1996, AAAI Press, 1996. "Station data", [online], [Search February 14, 2010], Internet <URL: http://www.ekidata.jp/index.html>

  While the latitude / longitude information of the current position of the mobile terminal can be acquired by GPS, it is impossible to acquire the meaning of the current position (whether or not it is a significant position). Even when the server provides service information corresponding to the position to the mobile terminal, the service information to be provided varies depending on the “meaning” (significance) at the position. For example, even if a place visits at a similar frequency, the contents of service information to be provided are completely different between a favorite store and a business partner's office. Significant positions vary depending on the individual. Among them, the “address” like home and the “residence” like work / school are considered the most significant positions. Even if it is possible to extract “a place where the stay frequency is high” by using the technique described in Patent Document 1, it is not possible to specify the significance thereof. Usually, the “semantic assignment” for the positioning position is specified by the user, or other registration information needs to be used.

  Therefore, an object of the present invention is to provide a portable terminal, a server, a program, and a method for automatically estimating an address / location with a user's movement using only a positioning function mounted on the portable terminal.

According to the present invention, a portable terminal possessed by a user's movement, a positioning unit that receives positioning radio waves of GPS (Global Positioning System) , a positioning unit that is activated at a predetermined time period, and a positioning position is determined. A portable terminal having position acquisition means for acquiring and position storage means for storing a positioning position according to time,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for counting the number of stay days where the stay position exists even once for each cluster based on the day attribute in order to absorb the instability of GPS positioning ease ;
It is characterized by having a stay characteristic estimating means for estimating the stay characteristic of the cluster based on the day attribute in order from the cluster having the longest stay days.

According to another embodiment of the mobile terminal of the present invention, the stay characteristic estimation means,
Day attributes are based on calendar or user-set weekdays and holidays,
Stay characteristics are based on "address" like home and "residence" like work / school,
Whether the first cluster with a high all-day stay rate is defined as an address and the second cluster with a high weekday stay rate other than the first cluster is defined as a residence,
Select two clusters with a high all-day stay rate and define one cluster with a high weekday stay rate as the residence and the other cluster as the address, or
It is also preferable to select two clusters having a high all-day stay rate, define one cluster having a high holiday stay rate as an address, and define the other cluster as a residence.

  According to another embodiment of the portable terminal of the present invention, a centroid position is calculated from a number of stay positions for the first cluster and the second cluster extracted by the stay characteristic estimation means, and the centroid of the first cluster is calculated. It is also preferable to further include a center-of-gravity position determining unit that defines the position as the “address” position and defines the center-of-gravity position of the second cluster as the “residence” position.

According to another embodiment of the mobile terminal of the present invention,
An area name storage means for storing the position in association with the area name;
It further includes name resolution means for searching for the “area name” closest to the address position and the residence location using the area name storage means, and defining the “area name” as the address at the address location and the residence at the residence location. It is also preferable.

According to another embodiment of the portable terminal of the present invention, the staying position extracting means is
Smoothing a plurality of positioning positions p _{t to} p _tn acquired within the past predetermined time T from the current time _t, and calculating a smoothed position pa _t at the current time t;
Calculating a displacement distance from the nearest past smoothed position pa _t-1 for smoothing position pa _t of the current time t,
Calculate the moving speed from the displacement time with respect to the displacement distance,
It is also preferable to extract the smoothed position pa where the moving speed is equal to or less than a predetermined threshold as the “stay position”.

  According to another embodiment of the portable terminal of the present invention, the position is a latitude / longitude vector, and the clustering means preferably uses DBSCAN (Density-Based Spatial Clustering).

According to the present invention, it is a server that can communicate with a portable terminal that is carried with the movement of a user,
The mobile terminal includes a positioning unit that receives GPS positioning radio waves, a position acquisition unit that activates the positioning unit at a predetermined time period and acquires a positioning position, a position storage unit that stores a positioning position according to time, a position A position transmission means for transmitting a large number of positioning positions stored in the storage means, and a stay characteristic estimation server for receiving a large number of positioning positions from the mobile terminal,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for counting the number of stay days where the stay position exists even once for each cluster based on the day attribute in order to absorb the instability of GPS positioning ease ;
It is characterized by having a stay characteristic estimating means for estimating the stay characteristic of the cluster based on the day attribute in order from the cluster having the longest stay days.

According to the present invention, a positioning unit that receives a positioning radio wave of GPS, a computer mounted on a portable terminal that is carried with the movement of a user, activates the positioning unit at a predetermined time period, and sets the positioning position. A program for functioning as a position acquisition means for acquiring and a position storage means for storing a positioning position according to time,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for counting the number of stay days where the stay position exists even once for each cluster based on the day attribute in order to absorb the instability of GPS positioning ease ;
The computer is made to function as a stay characteristic estimation means for estimating the stay characteristic of the cluster based on the day attribute in order from the cluster having the longest stay days.

According to the present invention, it is a portable terminal that has a positioning unit that receives GPS positioning radio waves and is possessed with the movement of the user, activates the positioning unit at a predetermined time period, acquires a positioning position, A method for estimating a user's address and whereabouts for a mobile terminal that accumulates positioning positions according to time,
A first step of extracting only a staying position other than the moving position according to the movement of the time-series positioning position;
A second step of dividing the plurality of residence positions into spatial clusters;
A third step of counting, based on the day attribute, the number of stay days in which a staying position exists even once for each cluster in order to absorb instability of GPS positioning ease ;
And a fourth step of estimating the stay characteristics of the cluster based on the day attribute in order from the cluster having the longest stay days.

  According to the mobile terminal, the server, the program, and the method of the present invention, it is possible to automatically estimate the address / location with the movement of the user using only the positioning function installed in the mobile terminal.

It is a spatial plot figure in the positioning position for every predetermined time period. It is a functional block diagram of the portable terminal in this invention. It is explanatory drawing which extracts the smoothing position from many positioning positions. It is explanatory drawing which extracts a residence position from several smoothing positions. It is explanatory drawing which divided the some residence position into the spatial cluster. It is explanatory drawing which counted the stay days for every cluster. It is explanatory drawing which calculates a gravity center position from many stay positions. It is a system block diagram which has a portable terminal and a stay characteristic server.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a spatial plot of positioning positions for each predetermined time period.

  A mobile terminal (for example, a mobile phone) possessed by a user's movement can acquire a position by receiving a positioning radio wave from a GPS satellite. According to FIG. 1, the user has a home address in “Fujimino City, Saitama Prefecture” and a work place in “Iidabashi, Minato-ku, Tokyo”. The user commute between home and work via Ikebukuro. Also, the user frequently moves to “Otemachi, Chiyoda-ku, Tokyo” as a visit destination. In such a case, the mobile terminal possessed by the user performs positioning at a predetermined time period (for example, every 15 minutes), thereby acquiring a large number of positioning positions corresponding to the time. The positioning position is represented as “latitude / longitude” when GPS is used.

  In the present invention, the “address” means the home of each user's life, for example, the location of his / her home which is the center of his / her life. Further, the “residence” means a place that is continuously connected but not so strong as to be the home of life, for example, the place of work or school.

  FIG. 2 is a functional configuration diagram of the mobile terminal according to the present invention.

  According to FIG. 2, the mobile terminal 1 includes a positioning unit 100, a position acquisition unit 111, a position storage unit 112, and an application processing unit 113. The positioning unit 100 receives a positioning radio wave transmitted from a GPS satellite and identifies the current position. The position acquisition unit 111 activates the positioning unit 100 at a predetermined time period and acquires the positioning position. The position accumulation unit 112 accumulates the positioning position according to the time. The application processing unit 113 executes various services according to the stay characteristics for the user estimated according to the present invention.

  Moreover, the portable terminal 1 in the present invention includes a stay position extraction unit 121, a clustering unit 122, a stay rate calculation unit 123, a stay characteristic estimation unit 124, a center of gravity position determination unit 125, a region name storage unit 126, And a name resolution unit 127. These functional components are realized by executing a program that causes a computer mounted on the mobile terminal to function.

[Residual position extraction unit]
The stay position extraction unit 121 extracts only stay positions other than the moving position according to the movement of the time-series positioning position. Specifically, the user removes a moving position in a car or a train and extracts a staying position at home or work.

  FIG. 3 is an explanatory diagram for extracting a smoothed position from a large number of positioning positions.

  According to FIG. 3A, a large number of positioning positions p (t) (t = 1 to 6) for each predetermined time period I are represented. Moreover, according to FIG.3 (b), the smoothing position pa (t) (t = 3-6) in the present time t is represented. The smoothed position pa (t) is obtained by smoothing a plurality of positioning positions p (t) to p (t−n) acquired within the latest past predetermined time T from the current time t. Since the positioning position obtained by GPS includes an error, the positioning position may be displaced even when it is not actually moved. Therefore, the error can be absorbed by calculating the smoothed position pa (t).

According to FIG.3 (b), it represents as follows.
I: predetermined time period T: last past predetermined time for smoothing p (t): positioning position at time t P (t) = vector (latt, lont)
n (t): Number of data that have been successfully measured within the most recent predetermined time T at time t
0 ≦ n (t) ≦ (T / I)
pa (t): When n (t)> 0, smoothing position at time t
pa (t) = (1 / n (t)) _ΣtT ^t p (t)

  FIG. 4 is an explanatory diagram for extracting a staying position from a plurality of smoothing positions.

According to FIG. 4A, the displacement distance from the most recent smoothing position pa (t-1) to the smoothing position pa (t) at the current time t is calculated. The displacement distance is calculated by the Euclidean distance as follows, for example.
d (pa (t), pa (t-1))
= √ ((lata (t) -lata (t-1)) ² + (lona (t) -lona (t-1)) ² )
d (a, b): Euclidean distance between points a and b

Next, the moving speed is calculated from the displacement time with respect to the displacement distance between the smoothed positions pa.
The moving speed is calculated by the following formula, for example.
v (t) = d (pa (t), pa (t-1)) / displacement time Displacement time: time difference between time t of pa (t) and time t-1 of pa (t-1)

  According to FIG. 4B, finally, only the smoothed position pa where the moving speed v (t) is equal to or less than the predetermined threshold value Vth is extracted. That is, the moving position is removed by removing the smoothed position where the moving speed v (t) is faster than the predetermined threshold value Vth. Then, it is possible to leave only the staying position where the moving speed is equal to or less than the predetermined threshold value.

[Clustering section]
The clustering unit 122 divides a group of a plurality of staying positions into zero or more spatial clusters.

  FIG. 5 is an explanatory diagram in which a plurality of stay positions are divided into spatial clusters.

  For example, according to FIG. 5, it is divided into four clusters. In order to cluster a plurality of staying positions, a known algorithm such as DBSCAN (Density-Based Spatial Clustering) (see, for example, Non-Patent Document 2) can be used. Although k-means may be used, the number k of clusters needs to be given in advance. On the other hand, DBSCAN is suitable in that a relatively intuitive cluster can be extracted without giving the number of clusters in advance.

  DBSCAN is a clustering method for extracting clusters having an arbitrary shape that is not hyperspherical, and derives directly density-rechable at two points. From a certain seed point, the density reachable relationship is directly traced, and the reachable maximal set is extracted as one cluster.

  According to DBSCAN, two parameters are used: Eps (epsilon, distance) and MinPts (minimum points). For example, Eps = 0.0005 and MinPts = 10 may be given in advance. Further, these parameters may be changed according to a predetermined time period of positioning or a result of final address location estimation.

[Stay rate calculator]
The stay rate calculation unit 123 calculates the stay rate based on the day attribute using the number of stay days where the stay position exists even once for each cluster.

  FIG. 6 is an explanatory diagram in which the number of stay days is counted for each cluster.

The stay rate calculation unit 123 counts the number of stay days in which the stay position exists even once for each cluster based on the day attributes (all days, weekdays, and holidays).
nd (c): Number of days staying all day in cluster c ndw (c): Days staying on weekdays in cluster c ndh (c): Days staying on holidays in cluster c

  Here, the reason for counting the staying position on a day basis will be described. According to the present invention, if the positioning position exists at least once a day, the stay days are incremented by one. In this case, even if the positioning position is small, the stay rate is 100% if the positioning position exists at least once every day. On the other hand, even if there are many positioning positions on any one specific day, the stay rate is low.

  The reason why the stay rate is not seen in the “number of positioning positions” is that, in the case of GPS, the number of positioning positions depends not on the length of the stay time but on the ease of positioning (ease of positioning). is there.

  Moreover, there are three reasons why the stay rate is not viewed as “stay time”.

  The first reason is that in order to derive the stay time, it is necessary to track the entry time and the entry time to the cluster, which increases the amount of calculation. The entrance time is the time when the positioning position is displaced from outside the cluster into the cluster. The appearance time is the time when the positioning position is displaced from the inside of the cluster to the outside of the cluster.

  The second reason is that the strictness of the staying time varies for each cluster due to the ease of positioning by GPS. That is, the stay time of a cluster that is easy to position is accurately determined, but the stay rate of a cluster that is difficult to position is extremely rough.

  The third reason is that the size of the cluster area easily affects the staying time. In other words, the larger the cluster, the longer the stay time. Therefore, the clustering accuracy affects the accuracy of the staying time. For example, the result of clustering for the same positioning position may be a cluster A in FIG. 6 or a cluster B depending on parameters. At this time, the stay time is longer in the cluster B having a larger cluster area, but the stay days are equal.

  For this reason, by calculating the stay rate on a day basis, it is possible to absorb the instability of GPS positioning ease and the instability of clustering accuracy and estimate the stay location with high accuracy. It becomes.

  Here, the definition of weekday / holiday may be set on Saturday and Sunday based on the calendar information, or may be set by the user according to the usage mode. When the mobile terminal is a mobile phone, a calendar / scheduler function installed as an application can be used.

Next, the stay rate calculating unit 123 calculates the stay rate from the stay days based on the day attribute for each cluster. The stay rate is calculated as follows, for example.
D: Days in all days Dw: Days in weekdays Dh: Days in holidays R (c): All-day stay rate in cluster c
R (c) = nd (c) / D
Rw (c): Weekday stay rate in cluster c
Rw (c) = ndw (c) / Dw
Rh (c): Holiday stay rate in cluster c
Rh (c) = ndh (c) / Dh

[Stay characteristic estimation unit]
The stay characteristic estimation unit 124 estimates a stay characteristic of a cluster having a high stay rate based on the day attribute. The stay characteristics are based on an “address” like home and a “residence” like work / school. For example, there are the following three estimation methods.

  First, regardless of weekdays and holidays, the first cluster with a high all-day stay rate (stay days / record days) is defined as “address”, and a second cluster with a high weekday stay rate other than the first cluster It may be defined as “location”.

Secondly, two clusters A and B having a high all-day stay rate may be selected, and “address / location” may be defined based on the following relationship based on weekdays.
When RwA> RwB: Cluster A “Address”, Cluster B “Location”
When RwA <RwB: Cluster A “Location”, Cluster B “Address”
When RwA = RwB: the cluster with the higher holiday stay rate is “address”

Third, two clusters A and B having a high all-day stay rate may be selected, and “address / location” may be defined based on the following relationship based on holidays.
When RhA> RhB: Cluster A “Address”, Cluster B “Location”
When RhA <RhB: Cluster A “Location”, Cluster B “Address”
When RhA = RhB: The cluster where the stay rate on weekdays is high is “resident”

  For example, according to FIG. 6, cluster 1 is defined as “address” and cluster 3 is defined as “resident”.

[Center of gravity position determination unit]
The center-of-gravity position determination unit 125 calculates the center-of-gravity position from a large number of stay positions for the first cluster and the second cluster extracted by the stay characteristic estimation unit 124.

  FIG. 7 is an explanatory diagram for calculating the center-of-gravity position from a large number of staying positions.

  By calculating the position of the center of gravity, it is possible to absorb the instability of the accuracy of clustering and estimate a relatively accurate address / location. Further, even if the area of the cluster is large, the center of gravity position is almost close. For example, the result of clustering for the same positioning position may be a cluster A in FIG. 7 or a cluster B depending on parameters. At this time, although the areas indicated by the clusters are different, the position of the center of gravity does not change greatly.

For the center of gravity position pc of a certain cluster c, an average value of a group of staying positions belonging to the cluster may be calculated. When the cluster c includes n _c staying positions (smoothing positions), the calculation is performed by the following formula.
pc _x = (1 / n _c ) _Σpai∈c pa _i

  Then, the center-of-gravity position determination unit 125 defines the center-of-gravity position of the first cluster as the “address” position and the center-of-gravity position of the second cluster as the “location” position.

[Regional name storage unit]
The area name storage unit 126 stores the position in association with the area name. For example, there is a database in which latitude and longitude are stored in association with the station name using the station name as an area name (see Non-Patent Document 3, for example). According to Non-Patent Document 3, for example, the following data is stored.
Railway overview code rr_cd Integer 2 digits Route code line_cd Integer 5 digits Railway code + Area code + Serial number Station code station_cd Integer 77 digits Route code + Serial number ・ ・ ・ ・
Railway summary name rr_name Character string 64 bytes
Line name line_name Character string 128 bytes
Station name station_name Character string 128 bytes
Prefecture code pref_cd Integer 2 digits Hokkaido-Okinawa + Other Latitude lat Number Longitude lon Number

[Name resolution section]
The name resolution unit 127 uses the region name storage unit 126 to search for the “region name” closest to the address position and the residence location, and defines the “region name” as the address of the address location and the residence of the residence location. . Specifically, when station data is used, the station name that has the latitude / longitude closest to the latitude / longitude of the address position is set as the area name of the address position. Further, the station name having the latitude / longitude closest to the latitude / longitude of the residence location is set as the regional name of the residence location. By defining the region name, it is possible to absorb instability of the gravity center position accuracy. These area names can be used in various ways by the application processing unit 113.

  For example, according to FIG. 7, the “Fujimino” station closest to the address location is defined as the area name of the address location. In addition, “Iidabashi” station closest to the residence location is defined as the area name of the residence location. Therefore, the “Suidobashi” station and the “Jimbocho” station far from the residence location are not selected as the regional names of the residence location. Even if the position of the center of gravity is deviated from the correct address position, the area name can be given the correct name.

  FIG. 8 is a system configuration diagram including a mobile terminal and a stay characteristic server.

  According to FIG. 8, the portable terminal 1 is possessed with a user's movement, and acquires a positioning position with a predetermined time period. The obtained positioning position is transmitted from the position information transmission unit 114 to the stay characteristic estimation server 2 via the network. The stay characteristic estimation server 2 has the functions of the stay point extraction unit 121 to the name resolution unit 127 described above. Thereby, the server can estimate the stay characteristic in the movement of the mobile terminal and transmit various service information to the mobile terminal.

  As described above in detail, according to the portable terminal, server, program and method of the present invention, the address / location associated with the movement of the user is automatically estimated using only the positioning function installed in the portable terminal. can do.

  Various changes, modifications, and omissions of the above-described various embodiments of the present invention can be easily made by those skilled in the art. The above description is merely an example, and is not intended to be restrictive. The invention is limited only as defined in the following claims and the equivalents thereto.

DESCRIPTION OF SYMBOLS 1 Mobile terminal 100 Positioning part 101 Wireless communication interface 102 Communication interface 111 Position acquisition part 112 Position storage part 113 Application processing part 114 Position information transmission part 115 Position information reception part 121 Staying position extraction part 122 Clustering part 123 Stay rate calculation part 124 Stay Characteristic estimation unit 125 Center of gravity position determination unit 126 Area name storage unit 127 Name resolution unit 2 Stay characteristic estimation server

Claims (9)

A portable terminal possessed by a user's movement, a positioning unit that receives GPS (Global Positioning System) positioning radio waves, and a position acquisition unit that activates the positioning unit at a predetermined time period and acquires a positioning position And a position storage means for storing the positioning position according to time,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for calculating a stay rate based on a day attribute using the stay days in which the stay position exists even once for each cluster in order to absorb instability of GPS positioning ease ;
A portable terminal comprising: a stay characteristic estimation unit that estimates a stay characteristic of a cluster with a high stay rate based on a day attribute. About the stay characteristic estimation means,
The day attribute is based on weekdays and holidays by calendar or user settings,
The stay characteristics are based on an “address” such as a home and a “residence” such as a workplace / school,
A first cluster having a high all-day stay rate is defined as the address, and a second cluster having a high weekday stay rate other than the first cluster is defined as the residence;
Select two clusters with a high all-day stay rate, define one cluster with a high weekday stay rate as the residence and the other cluster as the address, or
The portable terminal according to claim 1, wherein two clusters having a high all-day stay rate are selected, one cluster having a high holiday stay rate is defined as the address, and the other cluster is defined as the residence.   For the first cluster and the second cluster extracted by the stay characteristic estimation means, the center-of-gravity position is calculated from a large number of staying positions, the center-of-gravity position of the first cluster is defined as the “address” position, The mobile terminal according to claim 2, further comprising a center-of-gravity position determining unit that defines a center-of-gravity position of the cluster as a “location” position. An area name storage means for storing the position in association with the area name;
Using the area name storage means, search for the "area name" closest to the address position and the address location, and name resolution means for defining the "area name" as the address at the address position and the address location The mobile terminal according to claim 3, further comprising: The stay position extracting means includes
Smoothing a plurality of positioning positions p _{t to} p _tn acquired within the past predetermined time T from the current time _t, and calculating a smoothed position pa _t at the current time t;
Calculating a displacement distance from the nearest past smoothed position pa _t-1 for smoothing position pa _t of the current time t,
Calculate the moving speed from the displacement time with respect to the displacement distance,
The mobile terminal according to any one of claims 1 to 4, wherein a smoothed position pa where the moving speed is equal to or less than a predetermined threshold is extracted as a "staying position". The position is a latitude / longitude vector,
6. The mobile terminal according to claim 1, wherein the clustering unit uses DBSCAN (Density-Based Spatial Clustering). It is a server that can communicate with the mobile terminal that is possessed as the user moves,
The portable terminal includes a positioning unit that receives GPS positioning radio waves, a position acquisition unit that activates the positioning unit at a predetermined time period and acquires a positioning position, and a position storage unit that stores the positioning position according to time And a position transmission means for transmitting a large number of positioning positions stored in the position storage means, and a stay characteristic estimation server for receiving a large number of positioning positions from the mobile terminal,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for calculating a stay rate based on a day attribute using the stay days in which the stay position exists even once for each cluster in order to absorb instability of GPS positioning ease ;
A stay characteristic estimation server comprising: a stay characteristic estimation unit that estimates a stay characteristic of a cluster with a high stay rate based on a day attribute. A position acquisition means having a positioning unit for receiving GPS positioning radio waves, starting a computer mounted on a portable terminal carried with the movement of a user, and starting the positioning unit at a predetermined time period and acquiring a positioning position And a program for functioning as position storage means for storing the positioning position according to time,
A stay position extracting means for extracting only stay positions other than the moving position according to the movement of the time-series positioning position;
Clustering means for dividing a plurality of residence positions into spatial clusters;
A stay rate calculation means for calculating a stay rate based on a day attribute using the stay days in which the stay position exists even once for each cluster in order to absorb instability of GPS positioning ease ;
A program for a portable terminal, which causes a computer to function as a stay characteristic estimation unit that estimates a stay characteristic of a cluster with a high stay rate based on a day attribute. A portable terminal that has a positioning unit that receives GPS positioning radio waves, and is carried with the movement of the user, activates the positioning unit at a predetermined time period, acquires a positioning position, and A method for estimating a user's address and whereabouts for a mobile terminal that accumulates positioning positions,
A first step of extracting only a staying position other than the moving position according to the movement of the time-series positioning position;
A second step of dividing the plurality of residence positions into spatial clusters;
A third step of calculating a stay rate based on a day attribute using the stay days in which the staying position exists even once for each cluster in order to absorb instability of GPS positioning ease ;
And a fourth step of estimating a stay characteristic of the cluster for a cluster having a high stay rate based on the day attribute.

Images