JP2011164070A - Device and method for predicting visiting place - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device that accurately predicts a visiting place of a mobile terminal. <P>SOLUTION: A device 1 for predicting a visiting place includes a storage unit 22 that stores past travel routes of a mobile terminal, an area detection unit 11 that detects a travel of a mobile terminal, a travel route detection unit 12 that detects a new travel route of a mobile terminal, a first route determination unit 13 that determined whether or not the new travel route and the past travel route correspond to each other, and a visiting place prediction unit 16 that predicts a visiting place of a mobile terminal based on a determination result. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a visited place prediction apparatus and a visited place prediction method.

  In recent years, a system for predicting a visited location of a mobile terminal and distributing information related to the visited location has been developed. Patent Document 1 expresses a route traveled by a user in the past by a standard area mesh code, and predicts a next visited place using the route.

JP 2009-230514 A

  On the other hand, user behavior is also affected by the time of day. For example, you don't go from home to the office at night or on holidays. However, the prior art has not been able to predict the place to visit in consideration of time.

  Therefore, an object of the present invention is to predict a user's visited place more accurately by considering temporal identity.

  In order to solve the above-described problem, a visited site prediction apparatus according to the present invention includes a storage device that stores a past movement route composed of a plurality of pieces of area information of the mobile terminal and the number of movements of the past movement route, and a mobile terminal. Area detection means for detecting current area information of the mobile terminal at a predetermined cycle, movement path detection means for detecting a new movement route of the mobile terminal based on the area information detected by the area detection means, and the movement path detection means It is determined whether or not the new movement route detected by the above and the past movement route stored in the storage device correspond to each other based on the area information of the new movement route and the area information of the past movement route. Based on the first route determination means, the past movement route determined to be supported by the first route determination means, and the number of movements of the past movement route stored in the storage device A destination prediction means for calculating a route passage probability that the mobile terminal is passing through the past movement route and predicting a visit area after the current area of the mobile terminal based on the calculated route passage probability. The area information includes both an area through which the mobile terminal passes and a time at which the mobile terminal passes through the area information.

  The visited site prediction method of the present invention includes a storage step in which the visited site predicting apparatus stores a past travel route composed of a plurality of area information of the mobile terminal and the number of times of travel of the past travel route; An area detection step in which the device detects current area information of the mobile terminal in a predetermined cycle, and a visited area prediction device detects a new movement route of the mobile terminal based on the area information detected in the area detection step. Whether the travel route detection step and the visited site prediction apparatus correspond to the new travel route detected in the travel route detection step and the past travel route stored in the storage step, The first route determination step for determining based on the area information and the area information of the past travel route, and the visited place prediction device include the first route determination step. Based on the past movement route determined to be compatible and the number of movements of the past movement route stored in the storing step, the route passing probability that the mobile terminal is passing the past movement route is calculated. And a visited area prediction step for predicting a visited area after the current area of the mobile terminal based on the calculated route passing probability, and the area information includes an area through which the mobile terminal passes and a time through which the area passes. It is characterized by including both.

  According to the present invention, the visited site prediction apparatus can predict the visited location accurately by predicting the visited location based on the space, time, and number of times of travel in the past.

  Further, in the visited place prediction apparatus of the present invention, the first route determination means includes a new travel route based on an area included in the area information of the new travel route and an area included in the area information of the past travel route. A second route determination means for determining whether or not a past movement route corresponds; a time included in the area information of the new movement route and a past determined to be supported by the second route determination means; A third determination is made as to whether or not the new movement route corresponds to the past movement route determined to correspond by the second route determination means based on the time included in the area information of the movement route. And a route determination means.

  According to the present invention, the visited site prediction apparatus is more efficient by dividing the determination based on space and the determination based on time when determining the past movement route used for predicting the visited location. It is possible to accurately determine the past movement route. Therefore, it is possible to predict the visited place more efficiently and accurately.

  Further, in the visited place prediction apparatus of the present invention, the second route determination means includes one or more areas in which all of a predetermined number of areas included in the area information of the new travel route are included in the area information of the past travel route. It is preferable to determine whether or not the new travel route and the past travel route correspond to each other.

  According to the present invention, the visited site prediction device can more efficiently and accurately determine the past travel route when determining the past travel route used for predicting the visited location based on the space. it can. Therefore, it is possible to predict the visited place more efficiently and accurately.

  Further, in the visited site prediction apparatus of the present invention, the third route determination means is included in the current area information among the area information of the past travel routes determined to be supported by the second route determination means. The area information including an area having a short distance from the selected area information is selected in order, and whether or not there is a time included in the current area information between two times included in the selected two area information It is preferable to determine whether or not a movement route corresponds to a past movement route determined to correspond by the second route determination means.

  According to the present invention, the visited site prediction device can more efficiently and accurately determine the past travel route when determining the past travel route used for predicting the visited location based on time. it can. Therefore, it is possible to predict the visited place more efficiently and accurately.

  Further, in the visited site prediction apparatus of the present invention, the third route determination means sets a predetermined other time at a time obtained by subtracting a predetermined time from an earlier time of the two times and a later time. Depending on whether or not the time included in the current area information is between the added time and the new travel route, the past travel route determined to be supported by the second route determination means corresponds to It is preferable to determine whether or not.

  According to the above-described present invention, the visited area predicting device, when determining the past travel route used for predicting the visited place based on time, has a slight fluctuation in the passing time such as a train delay or a slightly earlier work. Can be absorbed, and the past movement route can be determined more efficiently and accurately. Therefore, it is possible to predict the visited place more efficiently and accurately.

  In the visited site prediction apparatus of the present invention, the visited site predicting unit corresponds to the number of movements of the past travel route determined to be supported by the second route determining unit and the third route determining unit. It is preferable to calculate the route passage probability based on the number of movements of the past movement route determined to be.

  According to the present invention, the visited site prediction apparatus can calculate the route passing probability more accurately. Therefore, it is possible to predict the visited place more accurately.

  Further, in the visited site prediction apparatus of the present invention, the visited site predicting means identifies an arbitrary plurality of areas as visited candidate areas, and corresponds to each visited candidate area by the first route determining means passing through the visited candidate areas. The past travel routes determined to be extracted are extracted, the route passage probabilities of each extracted past travel route are combined to calculate the area passage probability, and the visit candidate area with the largest area passage probability is set as the visit area It is preferable to predict.

  According to the present invention, the visited area prediction device can set an arrival area of a past travel route, an area associated with an advertisement or event information, and an arbitrary plurality of areas such as a station as a visit candidate area, It is possible to predict a more convenient place to visit.

  Further, in the visited site prediction apparatus of the present invention, the visited site predicting unit temporarily stops the prediction of the visited area when there is no past movement route determined to be supported by the first route determining unit. Then, it is preferable to resume the prediction of the visited area when a past movement route that is newly determined to correspond as a result of detecting a new movement route by the movement route detection means after stopping.

  According to the present invention, the visited site prediction apparatus starts the visited site prediction at the time when it is spatially overlapped with the past travel route, even if it is moving in a place that has never been visited in the past. It becomes possible.

  ADVANTAGE OF THE INVENTION According to this invention, the suitable information about a visited place and the area which will pass from now on can be provided.

It is a figure which shows the structure of the visited place prediction apparatus 1 which concerns on embodiment. It is a figure which shows the hardware constitutions of the visited place prediction apparatus 1 which concerns on embodiment. It is a figure which shows an example of the movement route information which concerns on embodiment. It is a figure which shows an example of the past movement path | route which concerns on embodiment. It is a flowchart for demonstrating operation | movement of the visited place prediction apparatus 1 which concerns on embodiment. It is a figure which shows an example of the area information which concerns on embodiment. It is a figure which shows an example of the new movement path | route which concerns on embodiment.

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

  FIG. 1 is a block diagram illustrating a configuration of a visited site prediction apparatus according to the present embodiment. As shown in FIG. 1, the visited area prediction apparatus 1 includes an area detection unit 11 (area detection unit), a movement route detection unit 12 (movement route detection unit), and a first route determination unit 13 (first route determination unit). , Second route determination unit 14 (second route determination unit), third route determination unit 15 (third route determination unit), visited site prediction unit 16 (visited site prediction unit), travel route passage probability calculation unit 17, area The passage probability calculation unit 18, the service providing unit 19, the display unit 20, the route information update unit 21, the storage unit 22 (storage device), and the movement route information storage unit 23 are configured.

  The visited site prediction apparatus 1 is configured by hardware such as a CPU. FIG. 2 is a hardware configuration diagram of the visited site prediction apparatus 1. As shown in FIG. 2, the visited area prediction apparatus 1 shown in FIG. 1 physically includes a CPU 51, a RAM 52 and a ROM 53 as main storage devices, an input device 54 such as a keyboard and mouse as input devices, a display, and the like. Output device 55, a communication module 56 which is a data transmission / reception device such as a network card, a positioning device 57 such as a position measuring device such as GPS, an auxiliary storage device 58 such as a hard disk, and the like. The function of each functional block shown in FIG. 1 is to read predetermined computer software (for example, a computer program for executing the processing of the flowchart shown in FIG. 5) on the hardware such as the CPU 51 and RAM 52 shown in FIG. Thus, the input device 54, the output device 55, the communication module 56, and the positioning device 57 are operated under the control of the CPU 51, and the data is read and written in the RAM 52 and the auxiliary storage device 58. Hereinafter, each functional block will be described based on the functional blocks shown in FIG.

  The storage unit 22 includes a movement route information storage unit 23. The data held in the storage unit 22 is physically held in the RAM 52 or the like. The storage unit 22 stores a past movement route composed of a plurality of area information of the mobile terminal and the number of movements of the past movement route. The area information includes both the area through which the mobile terminal passes and the time at which the mobile terminal passes through the area.

  As shown in FIG. 6, the movement route information 4 includes information (movement route information) about the movement route that the mobile terminal has moved in the past (ID41, departure area 42, arrival area 43, passage area 44, number of movements 45). Is registered for each movement route. The data registered in the ID 41 is identification information for specifying the movement route that the mobile terminal has moved in the past ("R1", "R2", etc., hereinafter referred to as the movement route R1, the movement route R2, etc.) It is data representing. The data registered in the departure area 42 includes area information (latitude, longitude, and measurement error) indicating the departure area (area including the start end (departure point) of the movement route) of each movement route registered in the ID 41. The same shall apply hereinafter. The data registered in the arrival area 43 is area information indicating the arrival area (area including the end point (arrival point) of the movement route) of each movement route registered in the ID 41. In addition, you may include the time which passed the area in each area information of the departure area 42 of the movement route information 4, the arrival area 43, and the passage area 44.

  The data registered in the passage area 44 is area information indicating the passage areas (excluding the departure area and the arrival area) that the mobile terminal has passed while moving along each movement route registered in the ID 41. The data registered in the number of movements 45 is data indicating the number of times the mobile terminal has moved along each movement route registered in the ID 41.

  The area detector 11 detects the current area information of the mobile terminal at a predetermined cycle. The area detection unit 11 is based on a position signal (such as a GPS signal indicating the current latitude and longitude of the mobile terminal) detected by the positioning device (for example, the current latitude / longitude of the mobile terminal). And a detection time thereof are detected every predetermined period (for example, every 5 minutes, every 10 minutes, etc.). The area detection unit 11 will be specifically described with reference to FIG. 7. The area detection unit 11 newly detects, for example, areas E21, E3, E4, and E11 (area where the mobile terminal exists) sequentially. Areas E5, E6, and E7 shown in FIG. 7 are areas detected in the past and are already registered in the movement route information 4 shown in FIG. 6, and areas E21 and E11 shown in FIG. 7 are detected in the past. This corresponds to the departure area E2 and the arrival area E1.

  The movement route detection unit 12 detects a new movement route of the mobile terminal based on the area information detected by the area detection unit 11. The movement path detection unit 12 detects the movement path along which the mobile terminal is moving based on the area (area where the mobile terminal exists) detected by the area detection unit 11. The movement route detection unit 12 will be described in detail with reference to FIG. 7. The movement route detection unit 12 is based on the areas E21, E3, E4, and E11 newly detected by the area detection unit 11. A new moving route R11 that is moving is detected. Here, the movement path detection unit 12 detects whether the mobile terminal is staying or moving as follows. When the area detection unit 11 sequentially detects the first area and the second area, the first and second areas overlap, and there is a common area shared by the first and second areas. When there is a common area as described above for the first, nth areas (n is a predetermined natural number) sequentially detected by the area detection unit 11, the area detection unit 11 is staying at the mobile terminal. Otherwise, it detects that the mobile terminal is moving.

  The first route determination unit 13 determines whether or not the new movement route detected by the movement route detection unit 12 and the past movement route stored in the storage unit 22 correspond to the area information of the new movement route. The determination is made based on the area information of the past movement route. The first route determination unit 13 includes a second route determination unit 14 and a third route determination unit 15. The determination of the first route determination unit 13 may be based on the determination of the second route determination unit 14 described later. The determination by the first route determination unit 13 may be based on the determination by the third route determination unit 15 described later. The determination by the first route determination unit 13 may be based on the determination by the second route determination unit 14 described later and the determination by the third route determination unit 15 described later.

  The second route determination unit 14 determines whether the new travel route and the past travel route correspond to each other based on the area included in the area information of the new travel route and the area included in the area information of the past travel route. Determine whether. The second route determination unit 14 determines whether or not the predetermined number of areas included in the area information of the new movement route overlap one or more areas included in the area information of the past movement route. It may be determined whether or not the past movement route corresponds. The second route determination unit 14 determines whether or not a new (current) travel route detected by the travel route detection unit 12 corresponds to a past travel route registered in the travel route information 4. The determination may be made based on the route information 4. More specifically, referring to FIG. 6 and FIG. 7, the second route determination unit 14 includes the past movement in which the new movement route R11 detected by the movement route detection unit 12 is registered in the movement route information 4. It is determined whether or not the route is supported.

  Here, the second route determination unit 14 will be described in detail. The second route determination unit 14 is registered in the movement route information 4 that is detected by the area detection unit 11 and that the mobile terminal has passed through the new movement route while moving and is located closest to the passage area. Whether or not there is a common area that overlaps a predetermined number N2 (for example, 2) of passing areas included in one past moving route is determined in order of passing for each passing area of the new moving route. Then, the second route determination unit 14 continuously determines that there is no common area as described above in a predetermined number N3 (for example, 2) or more of the plurality of passage areas of the new movement route. Except for this, it is determined that the new travel route corresponds to this past travel route registered in the travel route information 4.

  This will be specifically described with reference to FIGS. The second route determination unit 14 is in the position closest to the passage area (area E3) that is detected by the area detection unit 11 and the mobile terminal passes through the new movement route R11 while moving, and the movement route information. 4 includes a predetermined number N2 (= 2) passage areas (areas E5 and E6 are closest to the area E3 in distance) included in the past one movement route R1 registered in No. 4 (details). Includes a common area where the smallest rectangular area MBR1 (including area E5 and area E6) overlaps, and a passing area detected by the area detection unit 11 while the mobile terminal passes through the new movement route R11 (Area E4) and a predetermined number N2 (= 2) of communication routes R1 that are closest to the area E4 and registered in the movement route information 4 in the past. Whether or not there is a common area that overlaps the smallest rectangular area MBR2 (including areas E6 and E7 in detail) including areas (areas E6 and E7 are closest in distance from area E4), The determination is made for each of the areas E3 and E4. Since the area E3 is included in the MBR1 and the area E4 has a common area overlapping the MBR2, the second route determination unit 14 determines that all of the areas E3 and E4 have a common area. Therefore, the second route determination unit 14 determines that the new travel route R11 corresponds to the past travel route R1 including the areas E5, E6, and E7.

  The departure area of the new movement route determined by the second route determination unit 14 as corresponding to the past movement route registered in the movement route information 4 is the departure area or the arrival area of this past movement route. It corresponds to. Hereinafter, the two areas correspond to each other when the two areas overlap and there is a common area shared by the two areas. Specifically, referring to FIG. 7, the departure area (area E21) of the new travel route R11 corresponds to the departure area (area E2 “home” (FIG. 7)) of the past travel route R1. The second route determination unit 14 includes a departure area (area E2 “home” corresponding to the departure area (area E21) of the new movement route R11 among the plurality of past movement routes registered in the movement route information 4. ”(FIG. 7)) in the past is specified. Then, the second route determination unit 14 determines that there is a common region in which the area E3 that has passed first and the MBR1 that includes the passage areas (areas E5 and E6) of the movement route R1 are overlapped. It is determined that the mobile terminal is moving on the moving route R1. Next, if there is a common area where the area E4 that first passes after passing through the area E3 and the MBR2 that includes the passing areas (areas E6 and E7) of the movement route R1 overlap, the second route determination unit 14 Accordingly, it is determined that the mobile terminal R is moving even when it passes through the area E4 on the movement route R1 that is determined to be moving in the area E3 that passes immediately before passing through the area E4.

  When the second route determination unit 14 determines that the mobile terminal has changed from moving to staying, the route information update unit 21 corresponds to the new movement route and the past movement route registered in the movement route information 4. If it is determined by the second route determination unit 14 that it is, the area information (including each area information indicating the departure area, the arrival area, and the passage area of the new movement route) indicating this new movement route is In addition to adding to the area information indicating the past movement route registered in the movement route information 4, the number of movements included in this movement route information is increased by one and updated. When the second route determination unit 14 determines that the new travel route and the past travel route registered in the travel route information 4 do not correspond to each other, the route information update unit 21 selects the new travel route. The area information to be represented (including the area information indicating the departure area, arrival area, and passage area of the new movement route) is registered in the movement route information 4 as new movement route information.

  The route information update unit 21 will be specifically described with reference to FIG. When the second route determination unit 14 determines that the new movement route R11 and the past movement route R1 correspond to each other, the route information update unit 21 determines that the second route has been changed from moving to staying. If determined by the determination unit 14, the area information (including the area information indicating the departure area, the passage area, and the passage area of the new movement route R 11) representing the new movement route R 11 is included in the movement route information 4. It is added to the area information indicating R1.

  The route information update unit 21 registers the new movement route R12 and the movement route information 4 when the movement route detection unit 12 detects the new movement route R12 based on the areas E8 and E9 detected by the area detection unit 11. When the second route determination unit 14 determines that the past travel route that has been set does not correspond, information indicating the new travel route R12 (indicating the departure area, arrival area, and passage area of the new travel route R12) (Including each area information) is registered in the movement route information 4 as new movement route information. For example, the second route determination unit 14 is the smallest rectangle that includes an area E8 on the new travel route R12 and a predetermined number N2 (= 2) passage areas of the travel route R1 that is closest in distance from the area E8. There is no common area that overlaps the area (for example, MBR1), and the minimum including the area E9 on the new movement path R12 and the predetermined number N2 of passing areas of the movement path R1 that is closest in distance from the area E9. There is no common area that overlaps the rectangular area (for example, MBR1), and therefore there is no common area in a plurality of passing areas (areas E8 and E9) of the new movement route R12 that are equal to or greater than a predetermined number N3 (= 2). Therefore, it is determined that the new travel route R12 does not correspond to the past travel route R1 registered in the travel route information 4.

  When the route information is stored, the route information update unit 21 displays the route information, the passage number (the number of times of passage), the passage area and the passage date and time each time the route passes. May be stored as An example is shown in FIG.

The second path determination unit 14, when determining whether a new movement path and past movement path corresponds, its passage to the current position p _n of the past has passed through the same place movement path r _i and The number of times f _i (1 ≦ i ≦ I) may be specified. Specifically, when the second route determination unit 14 determines whether the new travel route and the past travel route correspond to each other, the past N positions including the present {p _{n−N + 1} ,..., P _n }, a past movement route r _i that partially overlaps and the number of times of passage f _i are specified.

  The third route determination unit 15 is based on the time included in the area information of the new travel route and the time included in the area information of the past travel route determined to be supported by the second route determination unit 14. Then, it is determined whether or not the new movement route corresponds to the past movement route determined to correspond to the second route determination unit 14. The third route determination unit 15 includes an area having a short distance from the area included in the current area information among the area information of the past movement route determined to be supported by the second route determination unit 14. Two area information is selected in order, and the new route and the second route determination unit 14 respond to whether there is a time included in the current area information between two times included in the two selected area information. It may be determined whether or not a past movement route determined to be in correspondence. The third route determination unit 15 uses the current area information between the time obtained by subtracting the predetermined time from the earlier of the two times and the time obtained by adding the predetermined other time to the later time. It may be determined whether or not the new travel route corresponds to the past travel route determined to be supported by the second route determination unit 14 depending on whether or not the time is included.

The third route determination unit 15 uses the current position information (p _n , t _n ) and the past movement route information (r _i , f _i ) (1 ≦ i ≦ I) specified by the second route determination unit 14 and The number of times f ′ _i that passed the same place on the same day / time is specified. Specifically, for r _i , first, the positions { ^i_m p ₁ ,..., ^I_m p _K } obtained when the route is passed m times (1 ≦ m ≦ f _i ) are extracted. Next, a set of position information constituting an area overlapping with _pn : ( ^i_m p _k1 , ^i_m t _k1 ), ( ^i_m p _k2 , ^i_m t _k2 ) is specified. Then, day of the week (weekdays, holidays) is determined to be temporally corresponding if they meet the ^{_{i_m t k1 -T ≦ t n ≦}} i_m t k2 + T identical. Here, T (minutes) is a constant for absorbing a slight fluctuation of the passing time (such as a train delay or a slightly early attendance). Note that the condition satisfied by t _n may be ^i_m t _k1 −T ₁ ≦ t _n ≦ ^i_m t _k2 + T ₂ . Here T ₁ and T ₂ are not equal. That is, the T swing width for error absorption may be asymmetric.

An example of the third route determination unit 15 is shown in FIG. passing number f ₁ of r ₁ is set to 2 times, for simplicity, both times and those passed to the same day as the current, and T = 0. The position acquired at the _first pass through r ₁ is { ^1_1 p ₁ , ^1_1 p ₃ , ^1_1 p ₅ }, and the position information that forms the overlapping area with the current position information (p ₁ , 9:01) is ( ^1_1 p ₁ , 9:00) and ( ^1_1 p ₃ , 9:05), it is determined that they correspond in time to satisfy 9: 00 ≦ 9: 01 ≦ 9: 05. Next, the position acquired at the second pass of r ₁ is { ^1_2 p ₁ , ^1_2 p ₂ , ^1_2 p ₄ , ^1_2 p ₅ }, and the position constituting the overlapping area with (p ₁ , 9:01) information _{^{(1_2 p 2, 21: 05}} ), (1_2 p 1, 21: 10) since it is, 21: 05 ≦ 9: 01 ≦ 21: for 10 does not satisfy the, if not temporally corresponding judge. As a result, the number of passages f ′ ₁ corresponding to time is one.

  Based on the past travel route determined to be supported by the first route determination unit 13 and the number of travels of the past travel route stored in the storage unit 22, the visited site prediction unit 16 A route passage probability that is passing through the past movement route is calculated, and a visit area after the current area of the mobile terminal is predicted based on the calculated route passage probability. The visited place prediction unit 16 determines the number of movements of the past travel route determined to be supported by the second route determination unit 14 and the past travel route determined to be supported by the third route determination unit 15. The route passage probability may be calculated based on the number of times of movement. The visited place prediction unit 16 identifies any plurality of areas as visit candidate areas, and the past travel routes determined to be supported by the first route determination unit 13 passing through the visit candidate areas for each visit candidate area May be extracted, and the area passage probability may be calculated by adding the route passage probabilities of the extracted past movement routes, and the visit candidate area having the largest area passage probability may be predicted as the visit area. When there is no past movement route determined to be supported by the first route determination unit 13, the visited site prediction unit 16 temporarily stops the prediction of the visit area, and after the stop, the movement route detection unit 12 As a result of detecting a new travel route, prediction of a visit area may be resumed when a past travel route determined to be newly supported appears.

  The visited site prediction unit 16 includes a movement route passage probability calculation unit 17 and an area passage probability calculation unit 18.

Movement path transmission probability calculation portion 17, the past of the respective movement path r _p, calculates the probability is passing through a r _p based on the spatial and temporal identity with p _n. At this time, due to the above-described decrease in the GPS positioning accuracy, there is a case where the route does not temporarily overlap with the route even though it is passing the past movement route. For this reason, if the route passage probability is calculated using only the current position information, an erroneous prediction is temporarily made. Therefore, in this patent, the movement path r _i determined that the second path determination unit 14 spatially corresponds to the past N positions {p _{n−N + 1} ,..., P _n } including the present. The path passage probability of r _i is calculated by Equation (1) from the number of times f _i (n) of f and f ′ _i (n) determined by the third path determination unit 15 to correspond in time. Here, I is the number of travel routes determined to be spatially supported by the second route determination unit 14, and α is a weighting factor. In addition, when the number I of travel routes determined to be spatially supported by the second route determination unit 14 becomes 0, when a past travel route corresponding spatially appears again, The visit site prediction process is restarted with N = 1.

  Α satisfies 0 ≦ α ≦ 1. In the case of α = 0, only a past route that matches spatially but does not match temporally is used. That is, when this place has passed in the past in a time zone different from this time, the place visited later is set as the prediction result. On the other hand, in the case of α = 1, only the past route that matches spatially and temporally is used. That is, when this place has passed in the past in the same time zone as this time, the place visited later is set as the prediction result. In this way, it is possible to eliminate the case where the past route happens to coincide spatially and temporally by reflecting the number of passes f of the travel route that does not coincide temporally but spatially coincident in the prediction result. it can.

The area passing probability calculation unit 18 specifies a moving route group that will pass through the area for the area p _i for which the area passing probability is to be calculated, and determines the total route passing probability of the moving route group as the area passing rate of the area Probability. Specifically, first, an area that is a calculation target of the area passing probability is extracted. This is preferably an arrival area of each movement route, an area associated with advertisement or event information, a station, or the like. Next, among the movement routes r _i (1 ≦ i ≦ I) calculated by the movement route passage probability calculation unit 17 that the route passage probability is greater than ^0, the route r ⁰ _{i ′} that spatially overlaps the area. (1 ≦ i ′ ≦ I ′) is specified. Then, the area passing probability of the area p _i is calculated by the following equation (2).

  The service providing unit 19 distributes event information and the like regarding these areas based on the area passing probability of each area. As the area passing probability is higher, it is preferable to distribute information preferentially, such as distributing more event information about the area or displaying it higher.

  The display unit 20 displays the service content output by the service providing unit 19 to the user using the output device 55 or the like.

  Next, the operation of the visited site prediction apparatus 1 will be described with reference to FIG. First, the current area is detected (S1). Specifically, the area detection unit 11 acquires current area information. Subsequently, the current movement route is detected (S2). Specifically, the movement path detection unit 12 detects the movement path from the area information acquired so far and the area information acquired this time. Subsequently, the same spatial route determination is performed (S3). Specifically, the second route determination unit 14 refers to a past travel route history and identifies a route that spatially overlaps the current travel route. Subsequently, the same temporal route determination is performed (S4). Specifically, with respect to the past travel route determined to be supported by the second route determination unit 14, the third route determination unit 15 includes a route temporally corresponding to the current travel route. , Get the number of passes.

  Subsequently, the route passage probability of the past movement route is calculated (S5). Specifically, the movement route passage probability calculation unit 17 determines the past movement route that has been spatially supported by the second route determination unit 14, the number of times of passage, and the third route determination unit 15. The past movement path determined to correspond spatially and temporally and the number of times of passage are acquired. Then, the route passage probability of each moving route is calculated by the equation (1).

Subsequently, the area passing probability of the specific area p _i is calculated (S6). Specifically, the area passing probability calculation unit 18 acquires the area p _i for which the area passing probability is calculated. Then, with respect to the movement route that is determined by the movement route passage probability calculation unit 17 to have a route passage probability greater than 0, a route that overlaps spatially with the area p _i is identified, and each area is expressed by the above equation (2). The area passing probability is calculated.

  Subsequently, an appropriate service is provided (S7). Specifically, based on each area and its area passing probability calculated by the area passing probability calculating unit 18, the service providing unit 19 specifies the area and the visiting area to be passed from now on, and information related to the area is displayed. To deliver.

  Finally, when the user stays in a specific area, the travel route information is updated (S8). Specifically, the route information update unit 21 uses the position information newly acquired this time to move to a past movement route determined to spatially correspond to the newly acquired movement route. Update the route.

  Next, the operation and effect of the visited site prediction apparatus 1 configured as described above will be described.

  The visited area prediction apparatus 1 of the present embodiment includes a storage unit 22 that stores a past movement route composed of a plurality of area information of the mobile terminal and the number of movements of the past movement route, and current area information of the mobile terminal. An area detection unit 11 that detects at a predetermined cycle, a movement route detection unit 12 that detects a new movement route of the mobile terminal based on the area information detected by the area detection unit 11, and a detection that is performed by the movement route detection unit 12 Whether or not the new travel route thus made corresponds to the past travel route stored in the storage unit 22 is determined based on the area information of the new travel route and the area information of the past travel route. 1 path determination unit 13, based on the past movement path determined to be supported by the first path determination unit 13 and the number of movements of the past movement path stored in the storage unit 22. A visited site predicting unit 16 that calculates a route passing probability that the terminal is passing through the past moving route, and predicts a visiting area after the current area of the mobile terminal based on the calculated route passing probability. The area information includes both an area through which the mobile terminal passes and a time at which the mobile terminal passes through the area information.

  According to the present embodiment, the visited place prediction apparatus can predict the visited place accurately by predicting the visited place based on the space, time, and number of times of movement in the past. In addition, it is possible to appropriately provide event information related to the visited place and train delay information used to go there.

  Further, in the visited place prediction apparatus 1 of the present embodiment, the first route determination unit 13 performs a new process based on the area included in the area information of the new travel route and the area included in the area information of the past travel route. The second route determination unit 14 that determines whether or not the travel route corresponds to the past travel route, and the time included in the area information of the new travel route corresponds to the second route determination unit 14. Whether or not the new movement route corresponds to the past movement route determined to be supported by the second route determination unit 14 based on the time included in the area information of the determined past movement route. And a third route determination unit 15 that determines whether or not.

  According to the present embodiment, the visited site prediction apparatus is more efficient by dividing a determination based on space and a determination based on time when determining a past movement route used for predicting the visited location. Can accurately determine the past movement route. Therefore, it is possible to predict the visited place more efficiently and accurately.

  Further, in the visited site prediction apparatus 1 of the present embodiment, the second route determination unit 14 is configured such that all the predetermined number of areas included in the area information of the new movement route are included in the area information of the past movement route. It is preferable to determine whether or not the new movement route corresponds to the past movement route depending on whether or not the above area overlaps.

  According to the present embodiment, the visited site prediction apparatus determines the past travel route more efficiently and accurately when determining the past travel route used for predicting the visited site based on the space. Can do. Therefore, it is possible to predict the visited place more efficiently and accurately.

  In the visited area prediction apparatus 1 according to the present embodiment, the third route determination unit 15 includes the current travel area information of the past travel routes determined to be supported by the second route determination unit 14. Two area information including an area having a short distance from the area included in the area information is selected in order, and whether there is a time included in the current area information between two times included in the two selected area information However, it is preferable to determine whether or not the new movement route corresponds to the past movement route determined to correspond to the second route determination unit 14.

  According to the present embodiment, the visited site prediction apparatus determines the past travel route more efficiently and accurately when determining the past travel route used for predicting the visited site based on time. Can do. Therefore, it is possible to predict the visited place more efficiently and accurately.

  Further, in the visited site prediction apparatus 1 of the present embodiment, the third route determination unit 15 performs a predetermined other operation at a time obtained by subtracting a predetermined time from an earlier time of the two times and a later time. A new movement route and a past movement route determined to be supported by the second route determination unit 14 depending on whether or not there is a time included in the current area information between the time and the time added to Is preferably determined.

  According to the present embodiment, the visited area predicting device determines the past travel route used for predicting the visited place based on the time, such as a train delay or a slightly earlier work time. The vibration can be absorbed, and the past movement path can be determined more efficiently and accurately. Therefore, it is possible to predict the visited place more efficiently and accurately.

  In the visited site prediction apparatus 1 of the present embodiment, the visited site predicting unit 16 includes the number of movements of the past moving route determined to be supported by the second route determining unit 14 and the third route. It is preferable to calculate the route passage probability based on the number of movements of the past movement route determined to be supported by the determination unit 15.

  According to the present embodiment, the visited site prediction apparatus can calculate the route passing probability more accurately. Therefore, it is possible to predict the visited place more accurately. In addition, due to GPS positioning error, even if you are moving on the past travel route, even if it does not temporarily overlap with the travel route, It is possible to estimate accurately.

  Further, in the visited site prediction apparatus 1 according to the present embodiment, the visited site predicting unit 16 specifies an arbitrary plurality of areas as visit candidate areas, and the first route determination passing through the visit candidate areas for each visit candidate area. The past travel routes determined to be supported by the unit 13 are extracted, the route passage probabilities of the extracted past travel routes are added together to calculate the area passage probability, and the visit candidate area having the maximum area passage probability Is preferably predicted as the visit area.

  According to the present embodiment, the visited site prediction device can set an arbitrary plurality of areas such as arrival areas of past travel routes, areas associated with advertisements and event information, and stations as candidate visit areas. This makes it possible to predict a more convenient place to visit.

  In the visited site prediction apparatus 1 according to the present embodiment, the visited site predicting unit 16 determines the visited area when there is no past movement route determined to be supported by the first route determining unit 13. The prediction is temporarily stopped, and after the stop, the travel route detection unit 12 detects a new travel route. As a result, when the past travel route determined to newly correspond appears, the prediction of the visited area is resumed. It is preferable to do.

  According to the present embodiment, the visited site prediction device 1 performs the visited site prediction at the time when it is spatially overlapped with the past travel route even if it is moving in a place that has never been visited in the past. It becomes possible to start.

DESCRIPTION OF SYMBOLS 1 ... Visit place prediction apparatus, 11 ... Area detection part, 12 ... Movement route detection part, 13 ... 1st route determination part, 14 ... 2nd route determination part, 15 ... 3rd route determination part, 16 ... Visit place prediction part , 17 ... Movement route passage probability calculation unit, 18 ... Area passage probability calculation unit, 19 ... Service providing unit, 20 ... Display unit, 21 ... Route information update unit, 22 ... Storage unit, 23 ... Movement route information storage unit

Claims (9)

A storage device for storing a past movement route composed of a plurality of area information of the mobile terminal and the number of movements of the past movement route;
Area detecting means for detecting current area information of the mobile terminal at a predetermined period;
A movement path detection means for detecting a new movement path of the mobile terminal based on the area information detected by the area detection means;
Whether the new movement path detected by the movement path detection means corresponds to the past movement path stored in the storage device, the area information of the new movement path and the area information of the past movement path, First route determination means for determining based on
The mobile terminal passes through the past movement route based on the past movement route determined to be supported by the first route determination means and the number of movements of the past movement route stored in the storage device. A visited place predicting means for calculating a route passing probability of the mobile terminal and predicting a visiting area after the current area of the mobile terminal based on the calculated route passing probability;
With
The area information includes both the area through which the mobile terminal passes and the time to pass through the area.
A visited place prediction apparatus characterized by that. The first route determination means includes
Second route determination for determining whether or not the new travel route corresponds to the past travel route based on the area included in the area information of the new travel route and the area included in the area information of the past travel route Means,
Based on the time included in the area information of the new travel route and the time included in the area information of the past travel route determined to be supported by the second route determination means, the new travel route and the second route Third route determination means for determining whether or not a past movement route determined to be supported by the determination means corresponds;
The visited site prediction apparatus according to claim 1, comprising: The second route determination means determines whether or not all the predetermined number of areas included in the area information of the new movement route overlap one or more areas included in the area information of the past movement route, Determine whether or not it corresponds to a past travel route,
The visited site prediction apparatus according to claim 2, wherein The third route determination unit includes an area having a short distance from an area included in the current area information among area information of past movement routes determined to be supported by the second route determination unit. Select two pieces of area information in order, and respond to the new movement route by the second route determination means depending on whether there is a time included in the current area information between two times included in the two selected area information To determine whether or not it corresponds to a past movement route determined to be
The visited site prediction apparatus according to claim 2 or 3, characterized in that The third route determination means includes current area information between a time obtained by subtracting a predetermined time from an earlier time of the two times and a time obtained by adding a predetermined other time to a later time. It is determined whether or not the new movement route corresponds to the past movement route determined to correspond by the second route determination means depending on whether or not there is a time included in the
The visited site prediction apparatus according to claim 4, wherein The visited place predicting means includes the number of movements of the past movement route determined to be supported by the second route determination means and the past movement route determined to be supported by the third route determination means. Calculating the route passage probability based on the number of movements of
The visited site prediction apparatus according to any one of claims 2 to 5, wherein The visited place predicting means identifies an arbitrary plurality of areas as visit candidate areas,
For each visit candidate area, the past movement routes determined to be supported by the first route determination means passing through the visit candidate area are extracted, and the route passage probabilities of the extracted past movement routes are added up to the area. Calculate the passing probability,
Predict the visit candidate area with the largest area passing probability as the visit area,
The visited site prediction apparatus according to any one of claims 1 to 6, wherein The visited place predicting means temporarily stops the prediction of the visited area when there is no past moving route determined to be supported by the first route determining means, and after the stop, the moving route detecting means As a result of detecting a new travel route, when a past travel route determined to be newly supported appears, the prediction of the visit area is resumed.
The visited site prediction apparatus according to any one of claims 1 to 7, wherein A visited place prediction method, wherein a visited place prediction device stores a past movement route composed of a plurality of area information of a mobile terminal and the number of movements of the past movement route;
The visited area prediction device detects the current area information of the mobile terminal at a predetermined cycle; and
A visited route predicting device detects a new moving route of the mobile terminal based on the area information detected in the area detecting step; and
The visited site prediction device determines whether the new travel route detected in the travel route detection step corresponds to the past travel route stored in the storage step, and whether the area information of the new travel route and the past travel A first route determination step for determining based on the area information of the route;
Based on the past travel route determined to be supported by the visited site prediction device in the first route determination step and the number of movements of the past travel route stored in the storage step, the mobile terminal A visit location prediction step for calculating a route passage probability that is passing through the travel route of the mobile terminal, and predicting a visit area after the current area of the mobile terminal based on the calculated route passage probability;
Including
The area information includes both the area through which the mobile terminal passes and the time to pass through the area.
A method for predicting a place to visit.

Images