JP2011128725A - Staying place estimation apparatus, method, and program - Google Patents

Abstract

Even when position data can be acquired only in a discrete and long cycle, it is possible to estimate a place where a user stays on a daily basis.
Based on GPS data acquired from mobile terminals MS1 to MSn, a location information generating unit 212 first defines a stay point and a destination point of the mobile terminals MS1 to MSn in a mesh form on map data. By local area. Next, the stay point calculation unit 213 calculates the stay probability for each area of the mobile terminals MS1 to MSn based on the information on the obtained stay point and the destination point, and further, the stay point reachability check unit 214 The movement probability for each time zone is calculated, and the transition of the place where the mobile terminal is staying is estimated from the movement probability for each time zone.
[Selection] Figure 3

Description

  The present invention relates to a stay location estimation apparatus, method, and program for estimating a point where a user stays in a daily action of the user.

As a technique for extracting a point where a user stays on a daily basis, for example, from a position data acquired by GPS (Global Positioning System), a time during which the user stays near a certain point is calculated, and the user has a threshold. If it stays in the vicinity of the point more than a value, the system which estimates the point as the staying place which has stayed is proposed. (For example, refer nonpatent literature 1.). There has also been proposed a method of extracting stay places that appear frequently on a daily basis and their order from a plurality of estimated stay places. Further, a method for estimating whether the direction obtained by GPS in a certain range is zero-crossing and estimating whether it is “staying” or “moving” has been proposed (for example, non-patent) See reference 2.)
These methods are based on the premise that position information can be acquired continuously, and can be implemented because position data can be acquired continuously in a short cycle with a GPS-dedicated device.

Nishino et al. (Nishino, Seko, Aoki, Yamada, Muto, Abe, “Examination of behavior pattern extraction method from stay location transition information”, IPSJ Research Report. UBI, 2008 (110) pp.57-64 G.C.De Silva, T.Yamasaki, K.Aiazawa, “Human Locomotion Segmentation by Constrained Hierarchical Clustering of GPS Data”, IPSJ Annual Conference, 21-2, Aug.27-29, 2008

  However, mobile terminals such as mobile phones and PDAs (Personal Digital Assistants) are not GPS-specific terminals, and have multiple sensors such as acceleration sensors and gyros as well as mobile communication functions and information processing functions. Yes. Therefore, for example, if the position data is continuously acquired by GPS with a short acquisition cycle, the power consumption is increased and the battery has only a few hours of life. In view of this, it is necessary to devise a technique for suppressing battery consumption such as increasing the GPS acquisition cycle on the application side. Further, as a characteristic of GPS itself, when the user is indoors such as a station or a train, position data cannot be acquired stably. Therefore, when using a mobile terminal, it is necessary to acquire position data in a discrete cycle having a long period of several minutes or more, and as a result, it is difficult to estimate the user's place of stay. .

  The present invention has been made paying attention to the above circumstances, and its purpose is to estimate the place where the user stays on a daily basis even when the position data can be acquired only in a discrete and long cycle It is in providing the stay location estimation apparatus, method, and program which were made.

  In order to achieve the above-mentioned object, one aspect of the present invention is that intermittently obtains position data of a mobile terminal from a mobile terminal that is an estimation target of a stay location, and the map data previously divided into a plurality of regions and the above-mentioned Based on the acquired position data, information representing the observation point of the portable terminal, its observation time, and the next destination point is obtained for each region and stored in a storage medium. Based on the accumulated observation point of the mobile terminal, the observation time thereof, and the next destination point, the stay probability for each area of the mobile terminal is calculated, and the calculated stay probability Based on the above, the staying point of the mobile terminal is extracted. Next, the movement probability of the portable terminal between the extracted stay points is obtained for each time zone, and the transition of the staying place of the portable terminal is estimated based on the movement probability of the portable terminal obtained for each time zone. It is what you do.

  Specifically, when extracting the above-mentioned staying points, first, the appearance of each of the above-mentioned observation points, based on the information indicating the accumulated observation point of the mobile terminal, its observation time, and the next destination point, respectively. The number of visits, the total observation time, and the number of stays where the next destination point is the same are calculated. Then, based on the calculated number of appearances, the total observation time, and the number of stays for each observation point, the probability that the mobile terminal stays at the observation point is calculated for each observation point. An area that is equal to or greater than a preset threshold value is extracted as a staying point of the mobile terminal.

  Further, based on the acquired position data of the mobile terminal, information representing the observation point for each time zone and the destination point for each time zone is further accumulated in the storage medium. In this state, when estimating the transition of the place of stay of the mobile terminal, staying for each predetermined time zone based on the accumulated observation points by time zone and destination points by time zone A matrix representing the connection between points is calculated for each combination of two different observation points by time zone, and based on the matrix showing the connection between residence points for each calculated time zone, the observation points by time zone are calculated. And a reachability matrix that represents whether or not the destination can be reached by time zone. And based on this calculated reachability matrix, while creating the stay point movement information by time zone representing the presence or absence of the movement from the observation point by the first time zone to the observation point by the second time zone, The movement probability of the mobile terminal for each observation point according to the time period based on the total stay point movement information according to the time period. Is calculated.

  Therefore, according to one aspect of the present invention, based on the position data acquired from the mobile terminal, the staying point and the destination point of the mobile terminal are obtained in units of regions defined on the map data, and this is obtained. The mobile terminal's stay probability for each region and the travel probability for each time zone are sequentially calculated based on the stay point and destination location, and the transition of the mobile device's stay location is estimated from the travel probability for each time zone. The For this reason, even if the position data of the portable terminal can be acquired only intermittently (discretely) due to the influence of the wireless environment of the portable terminal in order to suppress the power consumption of the portable terminal, It is possible to reliably estimate the place of stay.

  That is, according to the present invention, it is possible to provide a stay location estimation apparatus, method, and program capable of estimating a stay location where a user stays on a daily basis even when position data can be acquired only in a discrete and long cycle. Can be provided.

It is a figure which shows the structure of the system which implements the stay location estimation method concerning one Embodiment of this invention. It is a block diagram which shows the function structure of the portable terminal in the system shown in FIG. It is a block diagram which shows the function structure of the service server in the system shown in FIG. It is a flowchart which shows the process sequence and process content by the portable terminal shown in FIG. It is a flowchart which shows the process sequence and process content by the service server shown in FIG. It is a figure for demonstrating the search process of a point number among the processing procedures of the service server shown in FIG. It is a figure which shows an example of the positional information memorize | stored in the positional information storage part of the service server shown in FIG. It is a figure which shows an example of the stay probability in each point calculated by the stay point calculation part of the service server shown in FIG. It is a figure which shows an example of the staying point movement table according to the time slot | zone for one day produced | generated by the arrival check part between staying points of the service server shown in FIG. It is a figure which shows an example of the matrix showing the coupling | bonding between the daily residence points calculated by the arrival check part between residence points of the service server shown in FIG. It is a figure which shows an example of the reachability matrix computed by the reachable check part between staying points of the service server shown in FIG. It is a figure which shows an example of the stay point movement table classified by time zone which added the part for 5 days produced | generated by the arrival check part between stay points of the service server shown in FIG. It is a figure which shows an example of the movement probability for every residence point according to time zone produced | generated by the reachable check part between residence points of the service server shown in FIG.

Embodiments according to the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram showing a system including a stay location estimating apparatus according to an embodiment of the present invention. In this system, a plurality of mobile terminals MS1 to MSn each owned by a user can be connected to a service server SV as a stay location estimation device via a communication network NW.

  The communication network NW includes an IP (Internet Protocol) network and an access network for accessing the IP network. As the access network, an optical public communication network, a mobile phone network, a LAN (Local Area Network), a wireless LAN, a CATV (Cable Television) network, or the like is used.

The portable terminals MS0 to MSn are composed of a portable personal computer called a cellular phone, a PDA (Personal Digital Assistant), a netbook or the like, and is configured as follows. FIG. 2 is a block diagram showing the functional configuration.
That is, the mobile terminals MS1 to MSn include a control unit 11, a wireless transmission / reception unit 12, an encoding processing unit 13, a speaker 14 and a microphone 15 as an audio interface, a GPS (Global Positioning System) receiver 16, and a storage. A unit 17, an input unit 18, and a display unit 19 are provided.

  The wireless transmission / reception unit 12 transmits / receives a wireless signal to / from the communication network NW via the antenna 12a. The encoding processing unit 13 performs an encoding process and a decoding process necessary for transmitting and receiving information data such as voice and text and control data. The GPS receiver 16 receives GPS signals transmitted from a plurality of GPS satellites (not shown) via the antenna 16a. The storage unit 17 stores the information data and GPS data obtained by the GPS measurement unit 111 described later. The input unit 18 includes a dial keypad and a plurality of function keys, and is used by the user to input various information related to communication. The display unit 19 includes, for example, a liquid crystal display, and displays various information related to communication.

  The control unit 11 includes a GPS measurement unit 111 and a GPS data transmission control unit 112 as functions related to the present invention. The GPS measurement unit 111 activates the GPS receiver 16 at a predetermined measurement cycle of, for example, 5 minutes or more, and acquires the GPS data received by the GPS receiver 16 to obtain the GPS data of the terminal itself. The GPS data is composed of “time stamp”, “latitude”, and “longitude”. Then, the obtained GPS data is accumulated in the storage unit 17.

  When the GPS data transmission control unit 112 receives a GPS data transmission request from the service server SV, the GPS data transmission control unit 112 reads out the GPS data stored in the storage unit 17 and transmits it to the requesting service server SV. The GPS data may be read from the storage unit 17 and transmitted to the service server SV periodically or when the latitude and longitude of the terminal changes by a certain amount or more.

The service server SV is operated by, for example, a telecommunications carrier or a service provider, and is configured as follows. FIG. 3 is a block diagram showing the functional configuration.
That is, the service server SV includes a control unit 21, a communication interface 22, and a storage unit 23. The communication interface 22 transmits and receives information to and from the communication network NW under the control of the control unit 21.

The storage unit 23 is provided with a position information storage unit 231, a map database 232, and an estimation result storage unit 233 as storage units necessary for carrying out the present invention.
The location information storage unit 231 stores GPS data collected from the mobile terminals MS1 to MSn, that is, “time stamp”, “latitude”, and “longitude”, and “observation” generated by the location information generation unit 212 described later. Used for storing “time”, “point number”, “destination point number”, “time zone point number”, and “time zone destination point number”.

In the map database 232, latitude / longitude representing a position on the map and a point number defined by dividing the map into a plurality of points in a mesh form are stored in association with each other.
The estimation result storage unit 233 is used to store calculation results by a stay point calculation unit 213 and a stay point reachability check unit 214 described later.

  The control unit 21 has a central processing unit (CPU) as a core. As processing functions necessary for carrying out the present invention, a GPS data collection control unit 211, a position information generation unit 212, A stay point calculation unit 213 and a stay point reachability check unit 214 are provided. All of these processing functions are realized by causing the CPU to execute an application program.

  The GPS data collection control unit 211 periodically transmits a GPS data transmission request to each of the mobile terminals MS1 to MSn at different timings, and in response to this request, the GPS data transmitted from each of the mobile terminals MS1 to MSn is a communication interface. 22, and the process of storing in the position information storage unit 231 is executed.

The position information generation unit 212 executes the following processing.
(1) “Observed time” is calculated based on “time stamp” included in the received GPS data, and “point number” is calculated from the map database 232 based on “latitude” and “longitude”. And “movement destination point number” is searched. A process of storing the calculated and searched “observed time”, “point number”, and “destination point number” in the position information storage unit 231.
(2) From the “point number” and the “destination point number” stored in the position information storage unit 231 and the “time stamp”, the “point number by time zone” and the “destination point number by time zone” To obtain the position information and store it in the position information storage unit 231. The “time zone point number” is a point number for each time zone, and the “time zone destination number” is a destination point number for each time zone.

The staying point calculation unit 213 calculates a staying probability based on the position information stored in the position information storage unit 231 and obtains a staying point. Specifically, the staying point calculation unit 213 executes the following processing.
(1) Processing to calculate the “residence probability” for each location
First, based on the position information stored in the position information storage unit 231, the “number of observations”, “total observation time”, and “number of stays” are calculated for each point. Then, based on the calculated “number of observations”, “total observation time”, and “number of stays”, a “stay probability” is calculated for each point. However, “the number of observations” is the frequency at which a spot number appears in the position information stored in the position information storage unit 231, and “total observation time” is included in the position information stored in the position information storage unit 231. The total time for each “observed time” point, “the number of stays” is the same as the “location number” included in the location information stored in the location information storage unit 231. Of the frequency. Subsequently, based on the calculated “observation count”, “total observation time”, and “stay count”, the “retention probability” for each point is calculated.

(2) Processing to find the staying point
A point where the “Observation Count” calculated in (1) above is 1 or more is extracted, and the “Residence Probability” calculated in (1) above is a spot number that is equal to or greater than a predetermined threshold (referred to as a staying point) To extract.

  Next, the stay point reachability check unit 214 generates a stay point movement table for each time zone by obtaining a matrix and a reachable matrix representing the connection between the stay points for each day, and for each time zone for all days. The movement probability between the staying points by time zone is calculated from the staying point moving table. Specifically, the following processing is executed.

(1) A matrix C (retention point K_time zone & residence point L_time zone) representing the coupling between the residence points for each day is calculated for each combination of residence points by time zone. Here, the matrix C (the stay point K_time zone & the stay point L_time zone) representing the connection between the stay points is
(Number of point numbers by time zone other than stay points + two stay points by time zone)
* (Number of point numbers by time zone other than stay points + two stay points by time zone)
Each element cij is determined with reference to the “time zone point number” and the “time zone destination point number” stored in the position information storage unit 231.

  (2) Using the matrix C (retention point K_time zone & residence point L_time zone) representing the connection between the stay points calculated above, the time zone destination number and the time zone destination point number are Check if it is reachable. Then, it is checked whether it is reachable from the above calculated Mr (retention point K_time zone & residence point L_time zone).

(3) Using the above reachable matrix Mr (point K_time zone & location L_time zone), if moving from a certain time zone staying point to a certain time zone destination point number, 1 will not move If this is the case, a time-based staying point movement table that is 0 is created.
(4) First, for each period of data stored in the position information storage unit 231, the daily residence time table for each time zone is added up. Then, the movement probability Β (point K_time zone & location L_time zone) is calculated for each staying point by time zone.
Β (point K_time zone & location L_time zone)
= (Time zone point number is point K_time zone and time zone destination point number is point L_time zone value) / Σ (time zone point number is point K_time zone value).

Next, the stay location estimation operation by the system configured as described above will be described focusing on the operation of the service server SV.
First, in the mobile terminals MS1 to MSn possessed by each user, position information is detected and transmitted in the steady state as follows. FIG. 4 is a flowchart showing the processing procedure and processing contents.

  That is, the control unit 11 monitors whether or not the GPS measurement timing has come in step S41. For example, when 5 minutes have passed since the previous measurement timing, the GPS measurement unit 111 activates the GPS receiver 16 and takes in the GPS signal received by the GPS receiver 16 to calculate “latitude” and “longitude”. To do. Then, a “time stamp” indicating the measurement time is assigned to the calculated “latitude” and “longitude”, and “latitude” and “longitude” to which this “time stamp” is assigned are stored as GPS data in step S42. Stored in the unit 17.

  In addition, the control unit 111 monitors the reception of the GPS data transmission request in step S43 while periodically executing the GPS data measurement process. When a transmission request is received in this state, the process proceeds to step S44, and the GPS data transmission control unit 112 is activated. Then, the GPS data transmission control unit 112 reads out the GPS data stored after the previous transmission from the storage unit 17, and the read position information is read from the encoding processing unit 13 and the wireless transmission / reception unit 12 as a service server. Send to SV. Thereafter, every time a transmission request is received, the GPS data stored in the storage unit 17 is transmitted to the service server SV.

On the other hand, the service server SV executes processing according to the following procedure. FIG. 5 is a flowchart showing the processing procedure and processing contents.
That is, in step S51, the control unit 21 first monitors whether or not the GPS data collection control unit 211 has reached the GPS data transmission request timing. A GPS data transmission request is transmitted by division. When GPS data is sent from the mobile terminals MS1 to MSn, the GPS data is received in step S52, and the “time stamp”, “latitude” and “longitude” included in the GPS data are stored in the position information storage unit 231. Remember me.

  Next, the control unit 21 activates the position information generation unit 212 in step S53, and refers to the map database 232 under the control of the position information generation unit 212, such as “point number”, “destination point number”, and the like. The other elements constituting the position information are searched and stored in the position information storage unit 231.

For example, first, “observed time” is calculated based on “time stamp” included in the received GPS data. The calculation of this “observed time” is
“Observed time” = (time stamp at a certain time Tt) − (time stamp of GPS data obtained at the previous time Tt−1)
Is done.

  Further, based on “latitude” and “longitude” at a certain time Tt, a “point number” including a position represented by the “latitude” and “longitude” is searched from the map database 232. For example, if the map data is divided into a plurality of points in a mesh form as shown in FIG. 6, the numbers of these points are retrieved.

  Further, a “point number” including a position represented by “latitude” and “longitude” of the GPS data obtained at the next time Tt + 1 is searched for as a “destination point number”. Then, the calculated and searched “observed time”, “point number”, and “destination point number” are stored in the position information storage unit 231.

  Subsequently, the process proceeds to step S54, and from the searched "point number", "destination point number" and "time stamp", a "point number by time zone" representing a set of spot numbers for each time zone, The “destination point number by time zone” representing the set of destination point numbers for each time zone is obtained and stored in the position information storage unit 231.

  FIG. 7 shows an example of position information generated by the position information generation unit 212 and stored in the position information storage unit 231 as described above. In the figure, “1_10” of “point number by time zone” indicates that the user exists at point 1 in the time zone from 10:00 to 11:00, and “destination point number by time zone” “1_11” indicates that the user has moved to the point 1 in the time period from 11:00 to 12:00.

Next, the control unit 21 activates the staying point calculation unit 213, and calculates the staying probability and the staying point as described below under the control of the staying point calculation unit 213.
That is, the staying point calculation unit 213 first, in step S55, based on the position information stored in the position information storage unit 231, an “observation count” indicating the frequency of appearance of a certain spot number in the position information, Appearance when the “total observation time” that represents the total time of each “observed time” included in the location information is the same as the “location number” and the “destination location number” included in the location information The “retention frequency” representing the frequency is calculated for each point.

Then, based on the calculated “number of observations”, “total observation time”, and “number of stays”, a “stay probability” is calculated for each point. The following formula is used as the calculation formula.
"Residence probability" = F (total observation time) * G (number of observations, number of residences)
However,
F (total observation time) = If total observation time / number of observation days> α Then 1
Otherwise 0
G (number of observations, number of residences) = (number of residences) / (number of observations)
FIG. 8 shows an example of the values of “Observation Count”, “Total Observation Time”, “Retention Count”, and “Residence Probability” calculated for each of the points 1 to 16 with α = 60 minutes.

  Next, in step S56, the staying point calculation unit 213 first extracts a point where the calculated “number of observations” is 1 or more. In the example of FIG. 8, points 1, 6, and 16 are extracted. Then, a spot number with the calculated “staying probability” equal to or greater than a predetermined threshold is extracted and set as a staying point. For example, when the threshold value is set to “0.5”, the points 1 and 16 are extracted as the staying points in the example shown in FIG.

  Next, the control unit 21 starts the staying point reachability check unit 214, and under the control of the staying point reachability check unit 214, the matrix representing the connection between the staying points for each day and A reachable matrix is obtained to generate a time-dependent residence point movement table, and a movement probability between the time-zone residence points is calculated from the time-zone residence point movement table for all days.

That is, first, in step S57, a matrix C (retention point K_time zone & residence point L_time zone) representing the connection between the residence points for each day is calculated for each combination of residence points by time zone. Here, the matrix C (the stay point K_time zone & the stay point L_time zone) representing the connection between the stay points is
(Number of point numbers by time zone other than stay points + two stay points by time zone)
* (Number of point numbers by time zone other than stay points + two stay points by time zone)
Each element cij is determined as follows by referring to the “time zone point number” and the “time zone destination point number” stored in the position information storage unit 231.
cij = 1: When moving from the time zone point number m corresponding to the i row to the time zone destination point number n corresponding to the j row 0: Other than FIGS. 10 (a) and 10 (b) An example of a matrix C representing the coupling between the residence points when the residence points are 1_10 and 1_11 and a matrix C representing the coupling between the residence points when the residence points are 1_11 and 16_12 are shown.

Next, in step S58, by using the matrix C (the stay point K_time zone & the stay point L_time zone) representing the calculated connection between the stay points, the spot number by time zone and the destination by time zone Check whether the point numbers are reachable. For example, the reachable matrix Mr (point K_time zone & point L_time zone) is calculated based on the following Boolean algebra calculation rule.
Mr = (C + I) ^(p-1)
Here, p is the number of rows in the matrix C (retention point K_time zone & residence point L_time zone) representing the connection between the residence points. This determinant is described in “Humanoid Engineering”, Kumamoto, Tokyo Denki University Press, p48-51, P58-59.

Subsequently, whether or not it is reachable is checked by checking the following components of the calculated reachability matrix Mr (retention point K_time zone & residence point L_time zone).
When component = 1 (column number corresponding to staying point K_time zone, row number corresponding to staying point L_time zone) = 1:
From the stay point K_ time zone to the stay point L_ time zone, reachable without passing through other stay points.
When component = 0 (column number corresponding to staying point K_time zone, row number corresponding to staying point L_time zone) = 0:
The stay point L_ time zone is not reachable from the stay point K_ time zone.

FIGS. 11A and 11B show examples of reachability matrices when the stay points are 1_10 and 1_11, and reachability matrices when the stay points are 1_11 and 16_12, respectively.
Next, in the step S59, when the reachable matrix Mr (point K_time zone & location L_time zone) is used to move from a certain time zone staying point to a certain time zone destination point number, 1 is set. If it does not move, a time-based stay point moving table is created, which is 0. FIG. 9 shows an example of a time-dependent stay point movement table created on FIG. 7 for January 1, 2009.

Finally, in step S60, for each period of data stored in the position information storage unit 231, first, the daily residence time point movement table is added up. Then, the movement probability Β (point K_time zone & location L_time zone) is calculated for each staying point by time zone.
Β (point K_time zone & location L_time zone)
= (Time zone point number is point K_time zone and time zone destination point number is point L_time zone value) / Σ (time zone point number is point K_time zone value).
FIG. 12 shows an example of the time-dependent stay point movement table when the total for five days is added as described above. FIG. 13 shows an example of the movement probability 算出 calculated for each staying point by time zone based on the calculation result of the staying point moving table by time zone.

  As described above in detail, in this embodiment, based on the GPS data acquired from the mobile terminals MS1 to MSn, the location information generating unit 212 first determines the stay point and the destination point of the mobile terminals MS1 to MSn as map data. Obtained in units of regions defined in mesh form above. Next, the stay point calculation unit 213 calculates the stay probability for each area of the mobile terminals MS1 to MSn based on the information on the obtained stay point and the destination point, and further, the stay point reachability check unit 214 The movement probability for each time zone is calculated, and the transition of the place where the mobile terminal is staying is estimated from the movement probability for each time zone.

Therefore, the GPS data representing the positions of the mobile terminals MS1 to MSn is only intermittent (discrete) from the necessity of suppressing the power consumption of the mobile terminals MS1 to MSn or due to the influence of the wireless environment of the mobile terminals MS1 to MSn. Even if it cannot be acquired, it is possible to reliably estimate the place of stay of the mobile terminals MS1 to MSn in an arbitrary time zone.
By using the information on the estimated place of stay, it is possible to estimate, for example, an area or a store where the mobile user stopped without generating a blank section, and an advertisement or an advertisement corresponding to the estimated area or store Guidance and caution information can be distributed to the user. That is, an effective recommendation service can be implemented according to the user's place of stay.

The present invention is not limited to the above embodiment. For example, in the embodiment, the case where the GPS data collection control unit 211, the position information generation unit 212, the stay point calculation unit 213, and the stay point reachability check unit 214 are provided in the service server SV has been described as an example. However, the present invention is not limited to this, and each of the above functions may be provided in the mobile terminals MS1 to MSn so that the mobile terminals MS1 to MSn estimate the transition of the place where the mobile terminal itself or another mobile terminal stays.
In addition, the configuration of the GPS data collection control unit, the position information generation unit, the staying point calculation unit, and the staying point reachability check unit, and the processing procedure and processing contents thereof are variously modified without departing from the gist of the present invention. Can be implemented.

  In short, the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

  MS1 to MSn ... mobile terminal, SV ... service server, NW ... communication network, 11 ... mobile terminal control unit, 12 ... wireless transmission / reception unit, 13 ... encoding processing unit, 14 ... speaker, 15 ... microphone, 16 ... GPS reception 17 ... portable terminal storage unit, 18 ... input unit, 19 ... display unit, 111 ... GPS measurement unit, 112 ... GPS data transmission control unit, 21 ... service server control unit, 22 ... service server communication interface, 23 ... Storage unit of service server, 211 ... GPS data collection control unit, 212 ... Location information generation unit, 213 ... Residence point calculation unit, 214 ... Reachability check unit between residence points, 231 ... Location information storage unit, 232 ... Map Database, 233 ... estimation result storage unit.

Claims (7)

Means for intermittently obtaining position data of the mobile terminal from the mobile terminal to be estimated for the stay;
Based on the map data divided into a plurality of areas in advance and the acquired position data, information representing the observation point of the mobile terminal, its observation time, and the next destination point is obtained for each region. Means for storing in a storage medium;
Based on the accumulated observation point of the mobile terminal, the observation time thereof, and the next destination point, information indicating the stay probability for each region of the mobile terminal is calculated. And means for extracting a staying point of the mobile terminal;
Means for determining the movement probability of the mobile terminal between the extracted stay points for each time zone, and estimating a transition of the stay location of the mobile terminal based on the movement probability of the mobile terminal determined for each time zone; The stay place estimation apparatus characterized by comprising. The means for extracting the stay point is:
Based on the stored information indicating the observation point of the mobile terminal, the observation time, and the next destination point, the appearance count, the total observation time, and the next destination point are the same for each observation point. Means for calculating the number of stays, respectively,
Means for calculating the probability of the mobile terminal staying at the observation point for each observation point based on the calculated number of appearances for each observation point, the total observation time, and the number of stays;
The stay location estimation apparatus according to claim 1, further comprising means for extracting, as the stay point of the mobile terminal, an area where the calculated stay probability is equal to or greater than a preset threshold value. The means for accumulating further comprises means for accumulating in the storage medium information representing observation points by time zone and destination points by time zone based on the acquired position data,
The means for estimating the transition of the place of stay of the mobile terminal is:
A combination of two observation points for different time zones with a matrix representing a connection between dwell points for each predetermined time zone based on the accumulated observation points for each time zone and destination points for each time zone. Means for calculating each,
Based on the matrix representing the connection between the residence points for each calculated time zone, a reachability matrix is calculated that represents whether or not the observation point for each time zone and the destination point for each time zone can be reached. Means,
Based on the calculated reachability matrix, means for creating stay point movement information by time zone representing presence or absence of movement from the first observation point by time zone to the second observation point by time zone;
The travel point movement information by time zone is summed over all the accumulated observation periods, and the movement probability of the mobile terminal for each observation point by time zone based on the summed stay point movement information by time zone. The stay location estimation apparatus according to claim 1, further comprising: means for calculating The process of intermittently obtaining the location data of the mobile terminal from the mobile terminal that is the target of the stay,
The process of managing the observation point of the mobile terminal, the destination point and the observation time thereof in units of the area based on the map data previously divided into a plurality of areas, and the acquired position data,
Based on the managed observation point, destination point, and observation time of the mobile terminal, the probability of staying for each region of the mobile terminal is calculated, and the calculated threshold of staying is set in advance. A process of extracting an area that is equal to or greater than the value as a staying point of the mobile terminal;
Obtaining a movement probability of the portable terminal between the extracted staying points by time period, and estimating a transition of a staying place of the portable terminal based on the movement probability of the portable terminal obtained by the time period; The stay place estimation method characterized by comprising. The process of extracting the stay point is as follows:
Based on the stored information indicating the observation point of the mobile terminal, the observation time, and the next destination point, the appearance count, the total observation time, and the next destination point are the same for each observation point. The process of calculating the number of visits
A process of calculating the probability of the mobile terminal staying at the observation point for each observation point based on the calculated number of appearances for each observation point, the total observation time, and the number of stays;
The stay location estimation method according to claim 4, further comprising: extracting a region where the calculated stay probability is equal to or greater than a preset threshold as a stay point of the mobile terminal. The step of accumulating further comprises a step of accumulating in the storage medium information representing observation points by time zone and destination points by time zone based on the acquired position data,
The process of estimating the transition of the place of stay of the mobile terminal is as follows:
A combination of two observation points for different time zones with a matrix representing a connection between dwell points for each predetermined time zone based on the accumulated observation points for each time zone and destination points for each time zone. The process of calculating for each,
Based on the matrix representing the connection between the residence points for each calculated time zone, a reachability matrix is calculated that represents whether or not the observation point for each time zone and the destination point for each time zone can be reached. Process,
Based on the calculated reachability matrix, a process of creating stay point movement information by time zone that represents the presence or absence of movement from the first observation point by time zone to the second observation point by time zone;
The travel point movement information by time zone is summed over all the accumulated observation periods, and the movement probability of the mobile terminal for each observation point by time zone based on the summed stay point movement information by time zone. The stay location estimation method according to claim 1, further comprising:   The stay place estimation program which makes the computer with which the said stay place estimation apparatus performs the process corresponded to each means with which the stay place estimation apparatus in any one of Claims 1 thru | or 3 is provided.

Images