JP6607656B2 - Program, apparatus and method for estimating distance between staying places with a large amount of potential movement - Google Patents

Description

  The present invention relates to a technique for estimating a user's place of stay based on position information of a mobile terminal.

  Conventionally, there is a “person trip survey” to grasp the actual amount of movement of people in city planning. For example, it is possible to grasp “which transportation means”, “what kind of person”, “when” and “for what purpose” from the daily movement situation of an individual. As a result, it is possible to grasp the utilization ratio and traffic volume of transportation means (for example, railway, automobile, bus, bicycle, walking). Specifically, for a person who uses a predetermined target area, a destination, a purpose of movement, transportation, and a movement time zone are collected by a questionnaire. For this reason, it can be carried out only once every several years or when necessary, and when a new building or road is newly constructed, the influence of a disaster or the like cannot be reflected instantaneously.

  On the other hand, a technique for automatically measuring a movement amount using data measured by a measuring instrument or a ticket gate on a road and position information of a mobile terminal such as a smartphone has been attracting attention.

There is a technique for observing the amount of movement of each vehicle using a vehicle measuring instrument installed on a road (see, for example, Patent Document 1). According to this technique, OD (Origin, Destination) data in which a departure place, an arrival place, a time zone, and the number of vehicles are associated with each other is created, and a movement amount between the ODs is calculated.
There is also a technique for determining movement / staying based on position information of a mobile terminal and calculating a moving amount of a section longer than a certain staying time for each transportation facility (see, for example, Patent Document 2). According to this technique, it is possible to determine the transportation facility based on the moving speed.
Furthermore, there is a technique for generating a movable area of a mobile object based on position information of the mobile terminal (see, for example, Patent Document 3). According to this technique, the transportation facility is determined based on the acceleration, and the movement amount in each transportation facility is calculated based on the moving direction of the moving body.

JP 2005-182383 A JP 2005-115557 A JP 2006-066632 A

According to the technique described in Patent Document 1, it is necessary to install a vehicle measuring instrument on the road, and it is not intended to estimate an overall movement amount as in a person trip survey. In particular, the amount of movement cannot be measured for transportation other than cars and buses.
On the other hand, according to the techniques described in Patent Documents 2 and 3, it is possible to calculate the movement amount based on the OD between consecutive staying places.

  However, the inventors of the present application may have a large amount of potential movement between staying places that do not continuously transition in time series (no direct connection) due to the movement of the user. I thought. In other words, if we can estimate the amount of potential movement between places of stay, we may consider detouring between the places of stay and offloading the overall amount of movement of people. ? I thought.

  Therefore, an object of the present invention is to provide a program, an apparatus, and a method for estimating a distance between staying places with a large amount of potential movement.

According to the present invention, on the basis of time-series position information of a plurality of mobile terminals, a program that causes a computer to function to estimate a stay location pair with a large amount of potential movement between stay locations,
For each portable terminal U, a stay location estimation unit that determines a stay state from time-series position information and estimates the stay state position information as a stay location;
For each portable terminal U, a stay place group dividing means for dividing a group of stay places moved between predetermined points as one session;
A stay location pair extracting means for extracting all combinations of stay location pairs generated in the same session for each mobile terminal U;
A co-occurrence degree giving means for giving a co-occurrence degree based on the number of appearances of stay place pairs or stay places in all mobile terminals ΣU sessions for each stay place pair;
First stay place pair selection means for selecting only stay place pairs whose co-occurrence degree is equal to or greater than a predetermined threshold;
The computer is caused to function as second stay place pair selection means for selecting only stay place pairs that do not continuously transition in time series.

According to another embodiment of the program of the present invention,
The co-occurrence degree granting means is
For each mobile terminal U, the number of appearances = 1 for a staying place pair that appeared in the same session,
It is also preferable to make the computer function so that the sum of the number of appearances of the stay destination pairs in the sessions of all the mobile terminals ΣU is given as the co-occurrence degree.

According to another embodiment of the program of the present invention,
The co-occurrence degree granting means is
It is also preferable to cause the computer to function as a co-occurrence degree based on multiplication of the number of appearances of the first place of stay and the number of appearances of the second place of stay in all sessions of the mobile terminal ΣU.

According to another embodiment of the program of the present invention,
It is also preferable that the stay location estimating means extracts a stay for a predetermined time or more in the time-series position information, and causes the computer to function so as to estimate the stay position as the stay place.

According to another embodiment of the program of the present invention,
The place of stay estimation means estimates “home” from the time of stay for each mobile terminal,
It is also preferred that the staying place group sorting means causes the computer to function as a single session S for each portable terminal U, which is a group of staying places that start from the staying place at home.

According to another embodiment of the program of the present invention,
For each predetermined time range, a stay place estimation means, stay place group classification means, stay place pair extraction means, co-occurrence degree assigning means, first stay place pair selection means, and second stay place pair selection It is also preferred to have the computer function so that the means are executed.

According to the present invention, based on time-series position information of a plurality of mobile terminals, an apparatus for estimating a stay location pair with a large amount of potential movement between stay locations,
For each portable terminal U, a stay location estimation unit that determines a stay state from time-series position information and estimates the stay state position information as a stay location;
For each portable terminal U, a stay place group dividing means for dividing a group of stay places moved between predetermined points as one session;
A stay location pair extracting means for extracting all combinations of stay location pairs generated in the same session for each mobile terminal U;
A co-occurrence degree giving means for giving a co-occurrence degree based on the number of appearances of stay place pairs or stay places in all mobile terminals ΣU sessions for each stay place pair;
First stay place pair selection means for selecting only stay place pairs whose co-occurrence degree is equal to or greater than a predetermined threshold;
And second stay place pair selecting means for selecting only stay place pairs that do not continuously transition in time series.

According to the present invention, there is provided an estimation method for an apparatus for estimating a stay location pair having a large amount of potential movement between stay locations based on time-series position information of a plurality of mobile terminals,
The device
For each mobile terminal U, a first step of determining a stay state from time-series position information and estimating the stay state position information as a stay place;
A second step of dividing a group of staying places that have moved between predetermined points as one session for each mobile terminal U;
For each mobile terminal U, a third step of extracting all combinations of stay destination pairs that occurred in the same session;
A fourth step of giving a co-occurrence degree based on the number of appearances of the stay place pairs or stay places in all the mobile terminals ΣU sessions for each stay place pair;
A fifth step of selecting only the stay destination pairs whose co-occurrence is equal to or greater than a predetermined threshold;
And a sixth step of selecting only stay destination pairs that do not continuously transition in time series.

  According to the program, the apparatus, and the method of the present invention, it is possible to estimate the distance between staying places with a large amount of potential movement. Specifically, although there is no continuous transition, it is possible to estimate the distance between staying places where the amount of movement will be large. It is possible to consider detouring between such places and offloading the total amount of human movement.

It is a functional block diagram of the estimation apparatus in this invention. It is explanatory drawing showing the stay place transition of 1 session in the user U1. It is explanatory drawing showing the stay place transition of 1 session in the user U2. It is explanatory drawing showing the stay place transition of 1 session in the user U3. It is explanatory drawing showing the combination of all the stay place pairs for every user. It is explanatory drawing showing the co-occurrence degree based on the sum total of the appearance frequency of a stay place pair. It is explanatory drawing showing the co-occurrence degree based on the sum total of the appearance frequency of a stay place. FIG. 7 is a topology diagram in which the co-occurrence degrees in FIG. 6 are assigned to all stay destination pairs that continuously transition. FIG. 8 is a topology diagram in which the co-occurrence degrees in FIG. 7 are assigned to all stay destination pairs that continuously transition. FIG. 7 is a topology diagram in which the co-occurrence degrees in FIG. 6 are assigned to all stay destination pairs that do not transition continuously. FIG. 8 is a topology diagram in which the co-occurrence degrees in FIG. 7 are assigned to all stay destination pairs that do not continuously transition.

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

  FIG. 1 is a functional configuration diagram of an estimation apparatus according to the present invention.

The estimation apparatus 1 of the present invention can estimate staying place pairs with a large amount of potential movement between staying places based on time-series position information of a plurality of mobile terminals.
A mobile terminal 2 such as a smartphone or a tablet is always carried by the user, and sequentially transmits position information to the estimation device 1. The position information may be, for example, latitude and longitude measured by a GPS (Global Positioning System) mounted on the mobile terminal 2.

  According to FIG. 1, the estimation device 1 includes a stay place estimation unit 11, a stay place group classification unit 12, a stay place pair extraction unit 13, a co-occurrence degree giving unit 14, and a first stay place pair selection unit. 15 and a second stay destination pair selection unit 16. These functional components are realized by executing a program that causes a computer installed in the apparatus to function. Also, the processing flow of these functional components can be understood as an apparatus estimation method.

[Location Estimation Unit 11]
The stay location estimation unit 11 collects time-series position information measured by the mobile terminal 2. The position information may be received directly from the mobile terminal 2 or may be acquired from a specific position information server.

Next, the stay location estimation unit 11 determines the stay state from the time-series position information for each mobile terminal U, and estimates the position information of the stay state as “stay place”.
Specifically, the stay location estimation unit 11 extracts stays for a predetermined time (for example, 30 minutes) or more in the time-series position information, and estimates the stay location as “stay place”. A staying place ID (IDentifier) is given to each staying place.

  As another embodiment, the stay location estimation unit 11 may estimate “home” from a staying time zone for each mobile terminal. In order to estimate the home, for example, there are the following two methods, but other existing techniques may be used.

(Home estimation based on location information around the address)
The stay location estimation unit 11 registers in advance the address of the user who possesses the mobile terminal, and estimates the position information included in the predetermined radius centered on the address as “home”.
(Home estimation based on the tendency of residence time)
The stay location estimation unit 11 estimates whether or not the user is at home from the tendency of the time zone in which the user holding the mobile terminal stays.
For example, when the time-series position information is “for example, continuously for 6 hours or more in an area range having a radius of 100 m”, the area range can be estimated as “home”. Further, it is also preferable to further add a condition that “the day in which the user stays lasts 6 days or more in one week” and to estimate the area range as “home”.

[Stay Area Group Division 12]
The staying place group sorting unit 12 sorts a group of staying places that have moved between predetermined points as one session S for each portable terminal U.

  FIGS. 2, 3 and 4 are explanatory diagrams showing stay location transition of one session in the users U1, U2 and U3.

The “session” is divided into groups of staying places based on predetermined conditions for each portable terminal U.
For example, for each mobile terminal U, a group of staying places until the departure is made from the staying place estimated to be home (until leaving from the home and returning to the home) is classified as one session S. Also good. The session Sn exists independently for each user Uk. Of course, not only the place of stay at home, but also a start point and an end point may be set in advance and stay place transitions between them may be one session.
Moreover, as another embodiment, between A station and B station may be set as a session. As a result, it is possible to estimate the distance between the station A and the station B where there is a high potential movement amount and not between consecutive staying places.

According to FIG. 2, the staying place of the user U1 in one session is represented.
Continuous transition of the place of stay: A->B->C->D->E-> A
Location of one session: A, B, C, D, E
According to FIG. 3, the staying place of the user U2 in one session is represented.
Continuous transition of the place of stay: G->B->C->D->E->I->E->I->E->D->C->B-> G
Location of one session: B, C, D, E, I, G
According to FIG. 4, the place of stay of user U3 in one session is represented.
Continuous transition of the place of stay: J->E->I->E->D->F->D->C->D->E-> J
Location of one session: C, D, E, F, I, J

[Stay place pair extraction unit 13]
The stay place pair extraction unit 13 extracts, for each mobile terminal U, stay place pairs of all combinations generated in the same session. 2, 3, and 4 show the stay destination pairs of the users U1, U2, and U3.

FIG. 5 is an explanatory diagram illustrating combinations of all stay destination pairs for each user Uk.
According to FIG. 5, ten stay place pairs are extracted for one session Sk in the user U1, and fifteen stay place pairs are extracted for one session Sk in each of the users U2 and U3.

[Co-occurrence degree giving unit 14]
The co-occurrence degree giving unit 14 gives a co-occurrence degree based on the number of appearances of “stay place pairs” or “stay places” in the session S of all mobile terminals ΣU for each stay place pair. Specifically, there are the following two methods.
<Co-occurrence based on the sum of the number of appearances of staying destination pairs>
<Co-occurrence based on the total number of appearances of the place of stay>

<Co-occurrence based on the sum of the number of appearances of staying destination pairs>
For each portable terminal U, the number of appearances = 1 is set for the staying place pairs that appeared in the same session Sn. That is, the number of appearances = 1 is counted regardless of how many times the stay place pair ij that appeared in the same session Sn appears in the same session Sn. Specifically, the co-occurrence degree between the first stay place i and the second stay place j in the user Uk is expressed as follows.
pair (i, j, Uk, Sn) = 1
And the sum total of the appearance frequency of the stay place pair in all the sessions ΣS of all the mobile terminals ΣU is given as a co-occurrence degree.
pair (i, j) = Σ _Uk Σ _Sn pair (i, j, Uk, Sn)

FIG. 6 is an explanatory diagram showing the degree of co-occurrence based on the total number of appearances of stay destination pairs.
According to FIG. 6, the stay place pairs for each user are arranged in a column on the left side, and the total number of appearances for each stay place pair is shown on the right side as a co-occurrence degree.

<Co-occurrence based on the total number of appearances of the place of stay>
A numerical value based on multiplication of the number of appearances of the first place of stay and the number of appearances of the second place of stay in all the mobile terminal ΣU sessions is given as the co-occurrence degree. Specifically, it is expressed as follows.
pair (i, j) = √ (N (i) × N (j))
N (i): Number of appearances of stay i in all sessions ΣS of all mobile terminals ΣU N (j): Number of appearances of stay j in all sessions ΣS of all mobile terminals ΣU

FIG. 7 is an explanatory diagram showing the degree of co-occurrence based on the sum of the number of appearances of the stay place.
According to FIG. 7, the number of appearances of each staying place for each user is arranged in a column on the left side, and the number of appearances and the degree of co-occurrence for each staying place are shown on the right side. For example, the stay place N (B) = 3 represents that the number of appearances of the stay place B in all sessions ΣS of all mobile terminals ΣU is three.

[First stay destination pair selection unit 15]
The first stay place pair selection unit 15 selects only stay place pairs whose co-occurrence degree is equal to or greater than a predetermined threshold. Here, attention is paid to the apparent movement amount, and a stay place pair having a small co-occurrence degree has a small movement amount in the first place.

FIG. 8 is a topology diagram in which the co-occurrence degrees of FIG. 6 are assigned to all stay destination pairs that are continuously transitioned.
According to FIG. 8, only the stay destination pairs whose co-occurrence degree is equal to or greater than the predetermined threshold value 2 are selected. Here, only the following stay destination pairs are extracted.
B-C-D-E-I

FIG. 9 is a topology diagram in which the co-occurrence degrees of FIG. 7 are assigned to all stay destination pairs that continuously transition.
According to FIG. 9, only the stay location pairs whose co-occurrence degree is equal to or greater than the predetermined threshold value 4 are selected. Here, only the following stay destination pairs are extracted.
C-D-E-I

[Second place of stay pair selection unit 16]
The second stay place pair selection unit 16 selects only stay place pairs that do not continuously transition in time series. Whether or not the stay place continuously changes can be determined by the time series of the stay place estimated by the stay place estimation unit 11.

FIG. 10 is a topology diagram in which the co-occurrence degree of FIG. 6 is given to all stay destination pairs that do not continuously transition based on FIG.
According to FIG. 10, all the stay destination pairs in which the co-occurrence degree is equal to or greater than the predetermined threshold value 2 and does not continuously transition are represented with respect to the stay place pairs continuously transitioning in FIG. 8. This is because a staying place pair that actually continuously transitions in time series is not potential in the first place.
Here, the co-occurrence of the stay destination pair CE is the highest. That is, according to the embodiment based on FIG. 2 to FIG. 6 and FIG. 8, it can be understood that the amount of potential movement of the stay destination CE is the highest. For example, it is possible to consider a detour route for offloading the movement of people between the locations CE.

FIG. 11 is a topology diagram in which the co-occurrence degree of FIG. 7 is given to all stay destination pairs that do not continuously transition based on FIG.
According to FIG. 11, all the stay destination pairs in which the co-occurrence degree is equal to or greater than the predetermined threshold value 4 and does not continuously transition are represented with respect to the stay place pairs continuously transitioning in FIG. 9.
Here, the co-occurrence of the stay destination pair CE is the highest. That is, according to the embodiment based on FIGS. 2 to 5, 7, and 9, it can be understood that the amount of potential movement of the stay destination CE is the highest. For example, it is possible to consider a detour route for offloading the movement of people between the locations CE.

As another embodiment, for each “predetermined time range”, the stay place estimation unit 11, the stay place group classification unit 12, the stay place pair extraction unit 13, the co-occurrence degree giving unit 14, and the first stay place It is also preferable that the pair selection unit 15 and the second stay destination pair selection unit 16 are executed.
For example, “predetermined time range = 1 day” may be used. For example, if there are three times during the day when the user leaves the home and returns to the home (session), the number of sessions is three.

Further, the predetermined time range may be “a time period in which one day is divided”, “one week”, “one month”, or “seasonal unit”. In the predetermined time range, a plurality of sessions occur for the same user Uk.
pair (i, j, Uk) = Σ _Sn pair (i, j, Uk, Sn)
That is, it is possible to estimate the distance between places where there is a large amount of potential movement within a predetermined time range.

For example, when the predetermined time range is set to one month, the number of sessions is different for each user.
User U1-> number of sessions = 30 (Sn (n = 1 to 30))
User U2-> number of sessions = 40 times (Sn (n = 1 to 40))
User U3-> number of sessions = 42 (Sn (n = 1 to 42))

  As another embodiment, the process of the second stay place pair selection unit 16 is executed before the process of the first stay place pair selection unit 15 or the process of the co-occurrence degree giving unit 14. It can also be configured.

  As described above in detail, according to the program, the apparatus, and the method of the present invention, it is possible to estimate the distance between staying places with a large amount of potential movement. Specifically, although there is no continuous transition, it is possible to estimate the distance between staying places where the amount of movement will be large. It is possible to consider detouring between such places and offloading the total amount of human movement.

  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 Estimation apparatus 11 Stay place estimation part 12 Stay place group division part 13 Stay place pair extraction part 14 Co-occurrence degree provision part 15 1st stay place pair selection part 16 2nd stay place pair selection part 2 Portable terminal

Claims (8)

A program that causes a computer to function to estimate a stay location pair that has a large amount of potential movement between stay locations based on time-series position information of a plurality of mobile terminals,
For each portable terminal U, a stay location estimation unit that determines a stay state from time-series position information and estimates the stay state position information as a stay location;
For each portable terminal U, a stay place group dividing means for dividing a group of stay places moved between predetermined points as one session;
A stay location pair extracting means for extracting all combinations of stay location pairs generated in the same session for each mobile terminal U;
Co-occurrence degree giving means for giving a co-occurrence degree based on the number of appearances of stay place pairs or stay places in all mobile terminals ΣU sessions for each stay place pair;
First stay place pair selection means for selecting only stay place pairs in which the co-occurrence degree is equal to or greater than a predetermined threshold;
A program that causes a computer to function as second stay place pair selection means for selecting only stay place pairs that do not transition continuously in time series. The co-occurrence degree giving means is
For each mobile terminal U, the number of appearances = 1 for a staying place pair that appeared in the same session,
The program according to claim 1, wherein the computer is caused to function as a co-occurrence degree of the sum of the number of appearances of the stay destination pairs in the sessions of all the mobile terminals ΣU. The co-occurrence degree giving means is
Characterized in that the computer functions so as to give a numerical value based on multiplication of the number of appearances of the first place of stay and the number of appearances of the second place of stay in the sessions of all mobile terminals ΣU as the co-occurrence degree. The program according to claim 1. 4. The stay location estimation means extracts a stay of a predetermined time or more in time-series position information, and causes the computer to function to estimate the stay position as a stay location. The program according to item 1. The stay location estimation means estimates "home" from the time zone in which each mobile terminal stays,
The staying area group sorting means makes the computer function so as to classify a group of staying places as a single session S for each portable terminal U starting from the staying place at home. The program according to any one of claims 1 to 4. For each predetermined time range, the stay place estimation means, the stay place group classifying means, the stay place pair extracting means, the co-occurrence degree giving means, the first stay place pair selecting means, and the second 6. The program according to claim 1, wherein the computer is caused to function so as to execute the stay destination pair selection unit. Based on time-series position information of a plurality of mobile terminals, an apparatus for estimating a stay location pair with a large amount of potential movement between stay locations,
For each portable terminal U, a stay location estimation unit that determines a stay state from time-series position information and estimates the stay state position information as a stay location;
For each portable terminal U, a stay place group dividing means for dividing a group of stay places moved between predetermined points as one session;
A stay location pair extracting means for extracting all combinations of stay location pairs generated in the same session for each mobile terminal U;
Co-occurrence degree giving means for giving a co-occurrence degree based on the number of appearances of stay place pairs or stay places in all mobile terminals ΣU sessions for each stay place pair;
First stay place pair selection means for selecting only stay place pairs in which the co-occurrence degree is equal to or greater than a predetermined threshold;
And a second stay place pair selecting unit that selects only stay place pairs that do not continuously transition in time series. Based on time-series position information of a plurality of mobile terminals, an estimation method for an apparatus that estimates a stay location pair with a large amount of potential movement between stay locations,
The device is
For each mobile terminal U, a first step of determining a stay state from time-series position information and estimating the stay state position information as a stay place;
A second step of dividing a group of staying places that have moved between predetermined points as one session for each mobile terminal U;
For each mobile terminal U, a third step of extracting all combinations of stay destination pairs that occurred in the same session;
A fourth step of giving a co-occurrence degree based on the number of appearances of the stay place pairs or stay places in all the mobile terminals ΣU sessions for each stay place pair;
A fifth step of selecting only staying destination pairs in which the co-occurrence degree is equal to or greater than a predetermined threshold;
And a sixth step of selecting only stay destination pairs that do not continuously transition in time series.

Images