JP2014199212A - Navigation system, navigation method and navigation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve a navigation system capable of preferentially proposing a point to be highly useful for each user as a recommendation point.SOLUTION: A navigation system including a destination recommendation function comprises: a travel history database 43 for storing travel history data of a plurality of users between points in association with action patterns; a target action pattern determining section 26 for determining an action pattern suitable for an action tendency of a target user; a reference point setting section 27 for setting a guide reference point; a data extracting section 28 for extracting the travel history data T including the guide reference point from the travel history database 43 on the basis of a first extraction condition and a second extraction condition different from the first extraction condition; and a recommendation point presentation section 29 for presenting a first recommendation point and a second recommendation point determined by the travel history data T extracted on the basis of the first extraction condition and the second extraction condition.

Description

  The present invention relates to a navigation system that provides guidance information related to points that can be candidates for a destination. The present invention also relates to such a navigation method and navigation program.

  In the navigation system as described above, based on behavior history data collected from multiple users, a technology that recommends other destinations visited by other users who visited the same destination as recommended destinations (recommendation) Technology) is known. For example, an information providing apparatus described in International Publication No. 2004/075137 (Patent Document 1) records visit histories of a plurality of users, and exists in a section with a lot of user visits based on the recorded visit histories. It is configured to provide information about the facility. Patent Document 1 describes that it is possible to provide useful information even when a user visits an unfamiliar area by presenting a recommended place (recommended point) based on a similar behavior pattern of another person. Has been.

  However, in Patent Document 1, although it is sought to provide information based on “similar behavior patterns of others”, what is actually considered is a destination (several destinations visited by other users). Combination of destinations). That is, the “behavior pattern” in Patent Document 1 is used as a concept representing a simple history of a destination within a reference period (for example, one day). In this sense, the device of Patent Document 1 merely proposes other destinations having a high correlation with the designated destination as recommended points, for example, a user regarding a movement mode between the destinations. It is not intended to provide information that takes into account the behavioral trends (the “behavior pattern” referred to in the present application). For this reason, depending on the user, a place that is not very useful for the user because it does not match the behavior tendency may be suggested as the recommended spot.

International Publication No. 2004/075137

  Therefore, it is desired to realize a navigation system that can preferentially propose points that are likely to be beneficial to each user as recommended points.

  According to the present invention, a navigation system that provides guidance information related to points that can be candidates for a destination is defined by classifying a user's behavior tendency into a plurality of behavior patterns. A movement history database storing movement history data in which movement history is recorded in association with any one of the behavior patterns, and a target that matches the behavior tendency of the target user to be provided with the guidance information from the plurality of behavior patterns A target behavior pattern determination unit that determines a behavior pattern, a reference point setting unit that sets a guidance reference point according to a predetermined condition, and the movement history database that is associated with the target behavior pattern Based on each of the first extraction condition and the second extraction condition that is different from the first extraction condition, A data extraction unit that extracts the movement history data including an inner reference point, and a movement frequency between the movement reference data extracted from the movement history data by the data extraction unit based on the first extraction condition Presenting a relatively high point to the target user as a first recommended point, and among the movement history data extracted based on the second extraction condition by the data extraction unit, between the guidance reference point A recommended spot presenting unit that presents a spot with a relatively high movement frequency as the second recommended spot together with the first recommended spot.

  According to this feature configuration, the data extraction unit extracts data including the guidance reference point set according to a predetermined condition from the movement history data stored in the movement history database (first extraction). . Further, the data extraction unit extracts data associated with the target behavior pattern that matches the behavior tendency of the target user based on the first extraction condition (second extraction). The movement history data finally extracted through both of these includes information on other points visited by other users who match the behavior tendency of the target user when they visited the guide reference point in the past. It will be a thing. Among the plurality of points included in such movement history data, a point with a relatively high movement frequency with the guide reference point is set as the first recommended point, which may match the intention of the target user. A high spot can be preferentially proposed as the first recommended spot. Therefore, information that is highly likely to be useful can be provided to the target user.

  Moreover, according to said characteristic structure, movement history data are extracted based on the 2nd extraction conditions which are conditions different from 1st extraction conditions (3rd extraction), and a point different from a 1st recommendation point is 1st. It can be presented as two recommended spots. Therefore, for example, as the second recommended point, another point that is relatively likely to match the intention of the target user as an auxiliary point of the first recommended point, or a completely different viewpoint from the first recommended point It is possible to present points that are surprising to the target user.

  Here, a user database storing home data, which is information indicating the home positions of the plurality of users, and the target user's home position and the guidance reference point from among a plurality of predetermined positional relationship patterns, A positional relationship pattern determining unit that determines a target positional relationship pattern that matches the positional relationship, and the first extraction condition includes a positional relationship pattern between the home position and the guidance reference point as the target positional relationship pattern. And the second extraction condition is that the positional relationship pattern between the home position and the guidance reference point is a different positional relationship pattern different from the target positional relationship pattern. Conditions are preferable.

  According to this configuration, since the first extraction condition includes, as a further condition, that the positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern, the first and second extraction conditions are included. The movement history data finally extracted through both of the above are the other users that match the behavior tendency of the target user, and the user whose positional relationship between the home position and the guidance reference point is close to the target user. This includes information on other points visited when the guide reference point is visited in the past. Therefore, it is possible to preferentially propose a point that is more likely to match the intention of the target user as the first recommended point. As a result, information that is highly likely to be useful can be provided to the target user.

  According to the above configuration, since the second extraction condition is that the positional relationship pattern between the home position and the guidance reference point is a different positional relationship pattern different from the target positional relationship pattern, The movement history data finally extracted through the third extraction is adjusted when a user whose positional relationship between the home position and the guidance reference point is different from the target user has visited the guidance reference point in the past. It includes information on other places visited. Therefore, a point that is unexpected for the target user, such as a point where the user is unlikely to visit due to the positional relationship between the home position and the guidance reference point, can be proposed as the second recommended point. As a result, it is possible to provide the target user with information that is likely to be useful, and at the same time, information that can evoke new interest.

  Further, it is preferable that the second extraction condition includes as a further condition that the movement history data is associated with the target action pattern.

  According to this configuration, since the second extraction condition includes, as a further condition, that the movement history data is associated with the target behavior pattern, the second extraction condition is finally extracted through the first and third extractions. The movement history data is obtained when another user who matches the behavior tendency of the target user and the user who has a different positional relationship between the home position and the guide reference point has visited the guide reference point in the past. It includes information on other points visited in accordance with. Therefore, although the position relationship between the home position and the guidance reference point is unexpected, a point that is highly likely to match the user's intention can be proposed as the second recommended point for the behavior pattern. As a result, it is possible to provide the target user with information that is likely to be useful, and at the same time, information that can evoke new interest.

  Further, the positional relationship pattern includes at least a relative azimuth between a point that can be the home position and a point that can be the guidance reference point, and a distance between the point that can be the home position and the point that can be the guidance reference point. It is preferable to classify into a plurality based on one side.

  According to this configuration, a plurality of users whose relative orientations between a point that can be a home position and a point that can be a guidance reference point are classified into the same positional relationship pattern, or a point that can be a home position A plurality of users who are close in distance to a guide reference point are classified into the same positional relationship pattern, or a plurality of users whose relative azimuth and distance are close are in the same positional relationship pattern. Can be classified. Therefore, the positional relationship between the home position and the guidance reference point can be appropriately classified, and appropriate movement history information can be extracted by the data extraction unit.

  In addition, the plurality of positional relationship patterns include a direction division formed by dividing a relative direction between a point that can be the home position and a point that can be the guidance reference point, a point that can be the home position, and the point It is defined on the basis of a combination with a distance section formed by dividing a distance from a point that can be a guide reference point into a plurality of distances, and the different positional relationship pattern includes the distance segment and the orientation segment that define the target positional relationship pattern It is preferable that any one of the distance section and the azimuth section is a positional relationship pattern that differs from the combination of.

  According to this configuration, a plurality of positional relationship patterns can be systematically classified based on the combination of the azimuth section and the distance section. In addition, for a combination of a distance section and an azimuth section that define the target position relation pattern, a different positional relation pattern is obtained by setting a positional relation pattern in which either the distance section or the azimuth section is different as a different positional relation pattern. Can be a positional relationship pattern that is relatively similar to the target positional relationship pattern. As a result, information that is highly likely to be useful can be provided to the target user.

  The data extraction unit extracts the movement history data including the guide reference point based on a third extraction condition different from the first extraction condition and the second extraction condition, and the third extraction condition is , On the condition that the movement history data is associated with another behavior pattern different from the target behavior pattern, the recommended spot presentation unit is extracted by the data extraction unit based on the third extraction condition In the movement history data, it is preferable that a point having a relatively high movement frequency with the guidance reference point is presented as the third recommended point together with the first recommended point and the second recommended point.

  According to this configuration, the data extraction unit extracts data associated with a behavior pattern different from the behavior tendency of the target user based on the third extraction condition (fourth extraction). The movement history data finally extracted through the first and fourth extractions is visited when another user who shows a behavior tendency different from the behavior tendency of the target user has visited the guidance reference point in the past. It includes information on other points. Among the plurality of points included in such movement history data, a point having a relatively high movement frequency with respect to the guidance reference point is set as the third recommended point, which is unexpected for the target user about the behavior tendency (pattern). A unique spot can be proposed as the third recommended spot. Therefore, while providing information that is highly likely to be useful to the target user, it is also possible to provide information that can arouse new interest.

  In addition, it is preferable that the third extraction condition includes as a further condition that the positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern.

  According to this configuration, the third extraction condition includes a further condition that the positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern. Therefore, the movement history data finally extracted through the first and fourth extractions is another user having a behavior tendency different from the behavior tendency of the target user, and the position of the home position and the guidance reference point When a user whose relationship is close to that of the target user has visited the guidance reference point in the past, information on other points visited is included. Thereby, although there is no unexpectedness for the target user in the positional relationship between the home position and the guidance reference point, a point having an unexpectedness regarding the behavior tendency (pattern) can be proposed as the third recommended point. As a result, it is possible to provide the target user with information that is likely to be useful, and at the same time, information that can evoke new interest.

  In addition, it is possible to evaluate in multiple stages, and the basic items for classifying each behavior pattern are defined as basic classification items, and the categories representing the respective evaluation stages of the basic classification items are defined as evaluation categories. The plurality of behavior patterns are respectively defined based on a plurality of evaluation categories for each of a plurality of basic classification items, and the different behavior pattern is a combination of evaluation categories for each basic classification item that defines the target behavior pattern On the other hand, it is preferable that the evaluation classification of one classification basic item is a different behavior pattern.

  According to this configuration, a plurality of behavior patterns can be systematically classified based on a combination of evaluation categories determined for each of a plurality of classification basic items. In addition, for each combination of evaluation categories of each basic classification item that defines the target behavior pattern, another behavior pattern is compared with the target behavior pattern by making an action pattern with a different evaluation category of one classification basic item a different behavior pattern. It becomes a similar behavior pattern. Thereby, it can suppress that the point greatly different from the intention in view of the behavior tendency of the target user is presented as the second recommended point. Therefore, it is possible to provide information with a relatively high possibility of arousing new interest.

  Examples of basic classification items include items related to the length of travel distance between destinations, the length of stay at the destination, and the start time or end time of a specific action set in advance. Based on the length of the moving distance between the destinations, the behavior pattern can be classified according to the width of the behavior range, for example. Further, based on the degree of the length of stay at the destination, the behavior pattern can be classified according to the length of time spent for a predetermined behavior, for example. Moreover, based on the start time or end time of specific action, an action pattern can be classified according to the timing which performs specific action, for example. Then, by combining a plurality of these, the action patterns can be classified more precisely.

  In addition, the plurality of basic classification items include at least a matter related to the length of the moving distance between the destinations, and the different behavior pattern is for a combination of evaluation categories of the basic classification items that define the target behavior pattern. It is preferable that the evaluation classification regarding the length of the moving distance between the destinations is different behavior patterns.

  According to this configuration, for example, for a user having a relatively wide action range, a popular spot located near the guidance reference spot can be presented as the second recommended spot. Therefore, the consciousness can be directed to such a user at a point existing at a short distance. On the other hand, for example, a highly popular point located far from the guidance reference point can be presented as a second recommended point to a user having a relatively narrow action range. Therefore, the consciousness can be directed to such a user at a point that exists at a long distance.

  Further, when a destination is designated by the target user, the reference point setting unit sets the designated destination as the guidance reference point, and the recommended point presenting unit includes the first recommended point and the recommended point It is preferable to present the second recommended point as a candidate for the next destination.

  According to this configuration, it is assumed that the target user provides guidance regarding the destination that he / she wants to visit, and there is a point that is highly likely to match the intention of the target user as a candidate for a subsequent visit point. Can be presented together with a point. Therefore, it is possible to provide additional useful information and information that can attract new interest while respecting the intention of the target user.

  The technical features of the navigation system according to the present invention having the above-described configurations can also be applied to a navigation method that provides guidance information related to a point that can be a destination candidate and such a navigation program. Therefore, the present invention can also cover such methods and programs. Note that a storage medium such as a DVD-ROM or a flash memory in which the navigation program is stored can also be a target of rights.

  In this case, the characteristic configuration of the navigation method is defined by classifying the user's behavior tendency into a plurality of behavior patterns, and moving history data in which movement histories between points of the plurality of users are recorded in any of the behavior patterns. A target behavior pattern determination step for determining a target behavior pattern that matches a behavior tendency of a target user to be provided with the guidance information from among the plurality of behavior patterns, using a movement history database stored in association; A reference point setting step for setting a guide reference point according to the determined conditions, a first extraction condition that is associated with the target action pattern from the movement history database, and a condition different from the first extraction condition Based on each of the second extraction conditions, the movement history data including the guidance reference point is A data extraction step to be performed, and a first recommendation for a point having a relatively high movement frequency between the guide reference point and the movement history data extracted based on the first extraction condition by the data extraction step A point that is presented to the target user as a point and has a relatively high movement frequency with the guide reference point among the movement history data extracted based on the second extraction condition in the data extraction step. And a recommended point presenting step of presenting with the first recommended point as the second recommended point.

  In this case, the feature configuration of the navigation program is defined by classifying the user's behavior tendency into a plurality of behavior patterns, and the movement history data in which movement histories between points of the plurality of users are recorded in any of the behavior patterns. A target behavior pattern determination function that refers to a movement history database stored in association with each other and determines a target behavior pattern that matches a behavior tendency of a target user to be provided with the guidance information from the plurality of behavior patterns; A reference point setting function for setting a guide reference point according to a given condition, a first extraction condition that is associated with the target action pattern from the movement history database, and a first extraction condition that is different Based on each of the second extraction conditions that are conditions, the movement history data including the guide reference point The first recommendation is a point having a relatively high movement frequency between the data extraction function to be issued and the movement history data extracted based on the first extraction condition by the data extraction function. A point that is presented to the target user as a point and has a relatively high movement frequency with the guide reference point among the movement history data extracted based on the second extraction condition by the data extraction function. The feature is that the computer realizes a recommended spot presenting function that is presented together with the first recommended spot as the second recommended spot.

  Naturally, these navigation methods and navigation programs can also obtain the effects of the navigation system described above. Furthermore, it is also possible to incorporate some additional techniques mentioned as examples of suitable configurations of the above-described navigation system into these navigation methods and navigation programs. In that case, the effect corresponding to each additional technique can be obtained.

Schematic diagram showing the overall configuration of the navigation system Block diagram showing schematic configuration of in-vehicle terminal device Block diagram showing the schematic configuration of the management server The figure which shows an example of action history data The figure which shows the data example of action history database The figure which shows an example of matching with each user and an action pattern Diagram showing an example of a behavior pattern determination method The figure which shows the data example of movement history database Conceptual diagram showing the correlation between points ascertained from the movement history database An example of a conceptual diagram showing the degree of correlation between points after the second filtering Another example of a conceptual diagram showing the degree of correlation between points after the second filtering The figure which shows an example of matching with each user and a positional relationship pattern The figure which shows an example of the direction division centering on the guidance reference point Schematic diagram showing the recommended point presentation method to the target user Flow chart showing processing procedure of action history database generation Flow chart showing processing procedure for moving history database generation The flowchart which shows the processing procedure of recommendation point presentation

1. Configuration of Navigation System An embodiment of the present invention will be described with reference to the drawings. As shown in FIG. 1, a navigation system 1 according to this embodiment includes an in-vehicle terminal device 2 such as a navigation device mounted on a vehicle, and a management server 5 provided to be able to communicate with the plurality of in-vehicle terminal devices 2. It is configured. In this navigation system 1, a plurality of in-vehicle terminal devices 2 are connected to a communication network 3 such as the Internet via a radio base station 4. Wireless communication between the in-vehicle terminal device 2 and the wireless base station 4 can be performed using, for example, a mobile phone network, a wireless LAN (Local Area Network), or the like. The management server 5 is also connected to the communication network 3.

  In the present embodiment, each in-vehicle terminal device 2 generates corresponding user action history data A, and transmits the action history data A to the management server 5 as a kind of probe information. The management server 5 extracts the movement history data T from the action history data A regarding each of a plurality of users collected from the plurality of in-vehicle terminal devices 2. The management server 5 stores the extracted movement history data T in association with each user's behavior pattern (behavior tendency classification result) and distributes the aggregated movement history data T to each in-vehicle terminal device 2. The in-vehicle terminal device 2 provides guidance information G related to a point that can be a destination candidate based on the received movement history data T while performing navigation processing such as route guidance to the destination.

  More specifically, for a target user (a user who is a target for providing guidance information G), a point where another user having the same or similar behavior pattern as the target user has visited in the past, and the target user's home A recommended destination is presented based on points visited by other users with the same or different positional relationship between the position and the destination. As will be described below, the navigation system 1 according to the present embodiment has a behavior pattern of the target user (behavior tendency regarding a movement mode between points) and a positional relationship pattern (a positional relationship between the user's home position and the guidance reference point). ) That is highly convenient and can be implemented as a destination recommendation function. In addition, a destination recommendation function that is surprising to the target user can be realized.

Note that the program (navigation program) of the navigation system 1 is stored in a storage medium such as a DVD-ROM or a flash memory and distributed to the user, or distributed from the network server to the user via the communication network. The distributed or distributed navigation program is stored in the storage device of each in-vehicle terminal device 2.

  FIG. 2 is a block diagram illustrating a schematic configuration of the in-vehicle terminal device 2. The in-vehicle terminal device 2 includes a local position determination unit 21, a navigation calculation unit 22, an action history data generation unit 23, an update processing unit 24, a communication control unit 25, a reference action pattern determination unit 26, a reference point setting unit 27, and data extraction. A unit 28, a recommended point presentation unit 29, and a positional relationship pattern determination unit 30. In each of these function units, a calculation unit for performing various processes on input data is configured by hardware, software (program), or both. The in-vehicle terminal device 2 is connected to a GPS receiver 31, an orientation sensor 32, a distance sensor 33, a calendar timer 34, a communication interface 35, a display input device 36, and an audio output device 37. The in-vehicle terminal device 2 is connected to a map database 41, an action history database 42, a movement history database 43, and a user database 44. These are configured to be able to exchange information with each other.

  FIG. 3 is a block diagram illustrating a schematic configuration of the management server 5. The management server 5 includes a communication control unit 51, a data collection unit 52, a behavior history database generation unit 53, a history behavior pattern determination unit 54, a movement history database generation unit 55, and a user database generation unit 56. In each of these function units, a calculation unit for performing various processes on the input data is configured by hardware, software (program), or both. The management server 5 is connected to the communication interface 58, the action history database 62, the movement history database 63, and the user database 64. These are configured to be able to exchange information with each other.

  The in-vehicle terminal device 2 and the management server 5 are the same as general-purpose computers such as arithmetic processing devices such as a CPU (Central Processing Unit), storage devices such as a hard disk and an optical disk, and temporary storage devices such as a RAM (Random Access Memory). The hardware configuration is configured.

  Map data M is stored (stored) in the map database 41 provided in the in-vehicle terminal device 2. The map database 41 stores information on the road network constituting the map data M. The map data M includes road network data including a plurality of nodes and a plurality of links corresponding to roads connecting the nodes. Further, the map data M includes link cost information for each link. The link cost is set according to the link length of each link and road attributes (speed limit, road type, etc.). The map data M is referred to by the navigation calculation unit 22 when executing map display processing, route search processing, and the like.

  The map database 41 also stores (stores) facility data F. The facility data F includes coordinate information indicating the position of each facility, attribute information indicating a name, a genre, and the like. Further, the facility data F may further include additional information about business hours, the presence or absence of a parking lot, and the like. The facility data F is referred to by the navigation calculation unit 22 when executing a destination search process or the like.

  The own position determination unit 21 is a functional unit that acquires own position information indicating the current position of the vehicle on which the in-vehicle terminal device 2 is mounted. The own position determination unit 21 is connected to a GPS (Global Positioning System) receiver 31. In the present embodiment, it is further connected to the direction sensor 32 and the distance sensor 33. The own position determination unit 21 performs a calculation for specifying the own position based on outputs from the GPS receiver 31, the direction sensor 32, and the distance sensor 33. When an imaging device such as a camera is mounted on the vehicle, it may be configured to perform self-localization with higher accuracy using an image recognition function.

  The navigation calculation unit 22 is a functional unit that executes various navigation processes such as self-location display, route search from the departure point to the destination, route guidance to the destination, and destination search. The navigation calculation unit 22 generates various images necessary for navigation processing such as a map image and a destination search screen, and performs processing such as displaying on the display input device 36. Further, the navigation calculation unit 22 also performs guidance by voice guidance or the like by the voice output device 37. The display input device 36 may be a device in which a display device such as a liquid crystal display and an input device such as a touch panel are integrated. The audio output device 37 can be constituted by a speaker or the like, for example.

  The behavior history data generation unit 23 is a functional unit that generates behavior history data A that records a behavior history of the in-vehicle terminal device 2 (that is, a user who operates a vehicle in which the in-vehicle terminal device 2 is mounted). The action history data generation unit 23 receives the own position information determined by the own position determination unit 21 and also receives information such as the year, month, day, and time provided from the calendar timer 34, and associates the two and records the action history. Data A is generated. In the present embodiment, the action history data A particularly includes information about the history of a point determined to have stayed for a predetermined time or longer. The action history data A can be information on the location history within a predetermined reference period (one day in this example).

  The stay longer than the predetermined time can be determined based on the fact that the own position indicated by the own position information does not move for a predetermined reference time (for example, 20 minutes). The length of the reference time can be appropriately set in consideration of a length that can be distinguished from a temporary stay for a simple break. When the destination is set by the user, it may be determined uniformly that the destination has stayed for a predetermined time or longer, regardless of the actual staying time. Hereinafter, a point determined to have stayed for a predetermined time or more and a point set as a destination are collectively referred to as a “visit point”. FIG. 4 shows an example of behavior history data A generated by the behavior history data generation unit 23. As shown in this figure, in this embodiment, each user's action history included in the action history data A includes a departure time from home, a return time, a point ID of a visited point, an arrival time at each visited point, and Contains information about the time spent at each visit.

  The action history data A also includes information on the position of each visited point (not shown). The position of each visited point is stored (stored) in advance in the map database 41 in association with the point ID, for example, and can be acquired from the map database 41 based on the point ID. Information on the position of each visit point is acquired as coordinate information specified by, for example, latitude and longitude. The action history data A also includes information on action contents at each visited point. Since the facility data F stored in the map database 41 includes attribute information indicating the genre of each facility, it is possible to determine the action content at each visited point based on the genre of each facility. For example, if a visit point belongs to a scenic spot or a shrine or temple, etc., “sightseeing”, “shopping” if it belongs to a supermarket or shopping mall, etc., “meal” if it belongs to a restaurant, etc. It is determined that it was done.

  The action history data A may further include information on a travel time between two visit points and a travel distance between the two visit points. In the present embodiment, these two additional pieces of information are obtained by calculation based on the information on the position of each visit point, the arrival time, and the stay time. Therefore, it may not be included in the action history data A.

  The update processing unit 24 is a functional unit that updates the behavior history database 42 by storing (storing) the behavior history data A generated by the behavior history data generation unit 23 in the behavior history database 42. The update processing unit 24 sequentially stores (stores) the behavior history data A generated by the behavior history data generation unit 23 in the behavior history database 42 and updates the behavior history database 42. The update processing unit 24 also plays a role of updating the movement history database 43 to the latest state when the movement history data T distributed from the management server 5 is received. Details of the movement history data T will be described later. Further, the update processing unit 24 has a function of generating and updating the user database 44. Specifically, when a user information registration operation including the home data K of the user is performed, the update processing unit 24 generates a user database 44 and stores the registered user information. When a change occurs in the user's address or the like, and the re-registration operation for the changed address or the like is performed, the update processing unit 24 updates the user database 44 based on the re-registered user information. Further, when receiving the user information of another user distributed from the management server 5, the update processing unit 24 updates the user database 44 to the latest state based on the received user information. The update of the movement history database 43 and the user database 44 may be either a differential update or a full update depending on the data format distributed from the management server 5.

  The user database 44 stores (stores) user information, which is information regarding various users, including home data K, which is information indicating the home positions of a plurality of users, for each user. The home data K is information for specifying the approximate location of the user's home location, and is, for example, a home address, a telephone number, a postal code, or the like. In the present embodiment, at the time of initial setting immediately after the purchase of the in-vehicle terminal device 2 (the vehicle in which the in-vehicle terminal device 2 is installed), the target user uses the display input device 36 or the like to enter a home address, a home phone number, or the like. The user information including the home data K of the target user is stored in the user database 44 by performing the user information registration operation. Further, by distributing user information of other users from the management server 5, user information including home data K of all users is stored in the user database 44 for each user ID (described later). In addition to the home data K, the user information may include various types of information such as places where the user frequently visits and favorite points.

  The communication control unit 25 is a functional unit that controls transmission / reception of information to / from the management server 5. The communication control unit 25 performs a process of transmitting the action history data A stored in the action history database 42 to the management server 5 at a predetermined timing. For example, the communication control unit 25 uses the communication interface 35 to transmit the action history data A generated on the day when the user who has boarded the vehicle on which the in-vehicle terminal device 2 is mounted returns home after finishing the action of the day. To the management server 5. As the communication interface 35, a dedicated communication module may be used, or a general-purpose communication device such as a mobile phone terminal may be used. In addition, when a user's action extends over a plurality of consecutive days, the action history data A of the day may be transmitted every time when the user arrives at an intermediate place of stay, or the action history data of each day is collected together when returning home. A may be transmitted. Whether or not the user has returned home can be determined based on the home position set in advance and the information on the own position specified by the own position determination unit 21. The same applies to the determination of whether or not the user has arrived at the accommodation.

  Each in-vehicle terminal device 2 is given a user ID for identifying each of a plurality of users (in-vehicle terminal device 2). When the in-vehicle terminal device 2 (communication control unit 25) transmits the action history data A, the user ID information is also transmitted.

  In addition, the communication control unit 25 performs processing for requesting the management server 5 to distribute the latest movement history data T. For example, the communication control unit 25 accesses the management server 5 periodically (for example, every week, every month, etc.), and requests distribution of the latest movement history data T at that time. Alternatively, the communication control unit 25 accesses the management server 5 at a timing manually specified by the user, and requests distribution of the latest movement history data T at that time.

  Further, when the user database 44 is generated and the user information of the registered target user is stored in the user database 44, the communication control unit 25 transmits the user information stored in the user database 44 to the management server 5. The process to do is also performed. When the user address is changed and the re-registered user information is stored in the user database 44, the communication control unit 25 transmits the re-registered user information to the management server 5. . The communication control unit 25 also transmits the user ID information together when transmitting the user information of the registered or re-registered target user. Further, the communication control unit 25 performs a process of requesting user information of each other user. For example, the communication control unit 25 accesses the management server 5 periodically (for example, every week, every month, etc.), and requests distribution of the latest user information at that time. Alternatively, the communication control unit 25 accesses the management server 5 at a timing manually designated by the user, and requests distribution of the latest user information at that time.

  The communication control unit 51 of the management server 5 is a functional unit that controls transmission and reception of information with each in-vehicle terminal device 2. The communication control unit 51 transmits the action history data A and the user information registered in each in-vehicle terminal device 2 from each in-vehicle terminal device 2, and the latest movement history data T and the user information of each user. When there is a distribution request, after establishing a connection with the in-vehicle terminal device 2, each data is transmitted / received via the communication interface 58.

  The data collection unit 52 is a functional unit that collects the action history data A of each user. As described above, when the in-vehicle terminal device 2 transmits the action history data A, the information of the user ID is also transmitted. The data collection unit 52 associates these user IDs with the action history data A, and collects the action history data A of a plurality of users in association with each user ID. Further, the data collection unit 52 has a function of collecting user information (including home data K) of each user in addition to the function of collecting the behavior history data A of each user. Further, when collecting the user information of each user, the data collection unit 52 collects the user ID transmitted together with the user information in association with each other.

  The user database generation unit 56 is a functional unit that generates and updates the user database 64. The user database generation unit 56 generates and updates the user database 64 by storing (storing) user information including the home data K of the plurality of users collected by the data collection unit 52 in the user database 64. The user database generation unit 56 stores user information of a plurality of users in the user database 64 in a format in which personal information cannot be specified in association with each user ID. When the registered home address is changed and the changed user information re-registered from each in-vehicle terminal device 2 (each user) is transmitted together with the user ID, the user database generating unit 56 The user information (including home data K) stored in the user database 64 is updated based on the user ID.

  The behavior history database generation unit 53 is a functional unit that generates and updates the behavior history database 62. The behavior history database generation unit 53 generates and updates the behavior history database 62 by storing (storing) the behavior history data A of a plurality of users collected by the data collection unit 52 in the behavior history database 62. The action history database generation unit 53 stores the action history data A of a plurality of users in the action history database 62 in association with each user ID (see FIG. 5).

  The history behavior pattern determination unit 54 is a functional unit that determines, from a plurality of behavior patterns, a behavior pattern that matches each behavior tendency of a plurality of users, based on the behavior history data A collected for each user. Here, the “behavior pattern” referred to in the present application is a concept representing an action tendency related to a movement mode between visit points and a stay tendency at each visit point. The history behavior pattern determination unit 54 classifies and defines user behavior trends into a plurality of behavior patterns, and alternatively determines a behavior pattern that matches each user behavior trend from among the plurality of defined behavior patterns. To do.

  The behavior pattern is based on at least one of the length of the travel distance between the visited points, the length of the staying time at the visited points, and a predetermined start time or end time of the specific action. It can be classified. In the present embodiment, as an example, behavior patterns are classified based on the distance traveled between the visit points, the stay time at the visit points, the departure time, and the lunch time (see FIG. 6). These are items that can be evaluated in a plurality of stages and are the basis for classifying each behavior pattern, and can be referred to as “classification basic items”.

  The history behavior pattern determination unit 54 evaluates the behavior tendency regarding the movement distance between the visit points by one of “short” / “standard” / “long”. In addition, the behavior tendency regarding the staying time at the visit point is evaluated by any one of “short” / “standard” / “long”. Further, the behavior tendency regarding the departure time is evaluated by any one of “early” / “standard” / “late”. Further, the behavior tendency regarding the lunch time is evaluated by any one of “early” / “standard” / “late”. These “standard” / “short” / “long” / “early” / “slow”, etc. are classifications representing the evaluation stages of the respective basic classification items, and can be referred to as “evaluation classifications”.

  The historical behavior pattern determination unit 54 evaluates each basic classification item using a statistical method. FIG. 7 schematically shows an evaluation method for the moving distance between the visited points. As shown in the upper part of FIG. 7, the history behavior pattern determination unit 54 refers to the movement distance included in the plurality of behavior history data A of the user who is paying attention, and calculates the frequency for each category according to the length. calculate. Next, as shown in the middle row, the history behavior pattern determination unit 54 deletes the estimated invalid data that may be irregular from all the data related to the movement distance, and extracts only the other estimated valid data To do. For example, a predetermined number of upper and / or lower data (for example, 5% each) when sorted in order of movement distance can be used as estimated invalid data.

  Next, as shown in the lower part, the history behavior pattern determination unit 54 calculates a representative value having the estimated effective data as a population. Examples of such representative values include an average value, a mode value, a median value, and the like, and in this example, an average value is calculated. Then, the history behavior pattern determination unit 54 compares the calculated representative value with at least one predetermined determination threshold (two in this example) and relates to the movement distance based on the magnitude relationship between them. The behavior tendency is determined by multi-level evaluation (three-level evaluation in this example). In addition, regarding the stay time at the visit point, the departure time, and the lunch time, the behavior tendency regarding each is determined by the same method.

  The history behavior pattern determination unit 54 classifies the behavior pattern for each user according to the combination of the evaluation categories for each basic classification item. In this example, the historical behavior pattern determination unit 54 performs three-level evaluation for each of the four classification basic items, and determines one behavior pattern from a total of 81 patterns defined by combinations of the evaluation categories. . In FIG. 6, as an example, the behavior patterns of the users specified by the user IDs “P1” to “P6” are “ABcc”, “BBbb”, “CCbc”, “AAaa”, “ACbc”, respectively. ”And“ CCba ”are shown.

  The movement history database generation unit 55 is a functional unit that generates and updates the movement history database 63. The movement history database generating unit 55 generates movement history database 63 by extracting movement history data T from action history data A of a plurality of users and storing (storing) this movement history data T in movement history database 63. Update. Here, the movement history data T is data in which movement histories between visiting points of a plurality of users (history of visiting points) are recorded. For example, referring to the action history data A of the “P1” user shown in the forefront of FIG. 5, it can be seen that the “P1” user visited each visit point in the order of D 1 → D 9 → D 2 → D 12. Based on this, the movement history database generating unit 55 moves the movement history data T representing that the “P1” user has moved from D1 to D9, moved from D9 to D2, and moved from D2 to D12, respectively. Is generated (see FIG. 8). The movement history database generation unit 55 similarly generates movement history data T for other users.

  The movement history database generation unit 55 stores the movement history data T for each user in the movement history database 43 in association with any of a plurality of defined behavior patterns. The movement history database generation unit 55 associates the user ID, the behavior pattern determined for each user by the history behavior pattern determination unit 54, and the movement history data T extracted from the behavior history data A with each other. 63. In the present embodiment, the movement history data T is generated as information including a user ID (see FIG. 8), and the behavior pattern of each user is stored in association with the user ID (see FIG. 6). Thereby, the action pattern of each user and the movement history data T are appropriately associated via the user ID. The movement history database generation unit 55 is a functional unit different from the behavior history database generation unit 53 described above, one of which is referred to as a first database generation unit and the other as a second database generation unit. It can also be called.

  The movement history data T of a plurality of users is used to calculate the correlation (the degree of correlation between points) between a plurality of visited points. FIG. 9 conceptually shows the degree of correlation between points ascertained from the movement history data T stored in the movement history database 63. In addition, what is shown here is the correlation between points based on all the user data. In this figure, a line connecting two points indicates that a user moving between the two points has existed significantly in the past. The thickness of each line indicates the degree of correlation, indicating that the correlation becomes higher (the past movement frequency was higher) as the line gets thicker, and the correlation becomes lower (the past movement frequency becomes lower). It was low). For example, in the example of FIG. 9, when attention is paid to the point D1, the point D1 has a very high correlation with the point D9, a high correlation with the point D2, and a certain degree of correlation with the point D4 is recognized. No significant correlation is observed with the point D6. Correlation between other points can be considered similarly. Based on the concept of the degree of correlation between points, the destination recommendation function is realized. This point will be described later.

  The management server 5 (communication control unit 51) distributes the movement history data T in accordance with the distribution request from each in-vehicle terminal device 2. The update processing unit 24 of the in-vehicle terminal device 2 newly generates the movement history database 43 based on the received movement history data T or updates the existing movement history database 43 to the latest state. In this sense, the movement history database 43 of each in-vehicle terminal device 2 is substantially a copy of the movement history database 63 of the management server 5, and the two databases 43 and 63 can be regarded as the same in the present invention. That is, the contents of both databases 43 and 63 are the same except for the deviation of the data contents due to the difference in update frequency. In this way, a movement history database 43 that is a substantial duplicate of the movement history database 63 generated in the management server 5 is generated in each in-vehicle terminal device 2.

  Moreover, the management server 5 (communication control part 51) distributes the user information of all the users stored in the user database 64 in the format which cannot specify personal information according to the distribution request from each vehicle-mounted terminal device 2. And the update process part 24 of the vehicle-mounted terminal device 2 updates the user database 44 to the newest state based on the user information of each user delivered from the management server 5 as mentioned above. In addition, the communication control unit 51 transmits the changed user information from each in-vehicle terminal device 2 (each user), and when the user database 64 is updated based on the changed user information, for example, the user The updated user information in the user database 64 is distributed to all the in-vehicle terminal devices 2 whose information is stored in the database 64, the in-vehicle terminal device 2 for which a user information distribution request has been issued even once. In this sense, the user database 44 of each in-vehicle terminal device 2 is substantially a duplicate of the user database 64 of the management server 5, and in the present invention, both databases 44 and 64 can be regarded as the same. That is, the contents of both databases 43 and 63 are the same except for the deviation of the data contents due to the difference in update frequency.

  The reference behavior pattern determination unit 26 of the in-vehicle terminal device 2 is a functional unit that determines a predetermined reference behavior pattern for the target user. The reference action pattern is an action pattern serving as a reference for execution of second to fourth filtering described later by the data extraction unit 28. In the present embodiment, such a reference behavior pattern includes a target behavior pattern and another behavior pattern. The target behavior pattern is a behavior pattern that matches the behavior tendency of the target user. Another behavior pattern is a behavior pattern different from the target behavior pattern.

  The reference behavior pattern determination unit 26 determines a target behavior pattern as a kind of reference behavior pattern. The reference behavior pattern determination unit 26 alternatively determines the target behavior pattern from the same plurality of behavior patterns defined by the history behavior pattern determination unit 54 of the management server 5. Various modes can be adopted as the target behavior pattern determination method by the reference behavior pattern determination unit 26. For example, the reference behavior pattern determination unit 26 may determine the target behavior pattern by a statistical method similar to or similar to the history behavior pattern determination unit 54 based on the behavior history data A stored in the behavior history database 42. good. Moreover, the reference | standard action pattern determination part 26 may acquire the information of the determination result by the log | history action pattern determination part 54 about the target user, and may determine a target action pattern based on it. Alternatively, the target action pattern may be determined by direct input by the target user manually. These may be changeable according to the accumulation amount of the action history data A for the target user. In the present embodiment, the reference behavior pattern determination unit 26 corresponds to the “target behavior pattern determination unit” in the present invention.

  Moreover, the reference | standard action pattern determination part 26 determines another action pattern as a kind of reference | standard action pattern. In the present embodiment, the reference behavior pattern determination unit 26 has one specific behavior pattern in which the evaluation category of a specific classification basic item is different from the combination of evaluation categories of each classification basic item that defines the target behavior pattern. Is determined as another behavior pattern. For example, for the “P3” user whose target behavior pattern is determined to be “CCbc” in the example of FIG. 6, as an example, the evaluation category of “movement distance” that is one of four basic classification items is “long”. “ACbc” forcibly changed from “short” to “short” is determined as another action pattern. Alternatively, “CCba” in which the evaluation category of “lunch time”, which is one of the four basic classification items, is forcibly changed from “late” to “early” is determined as another action pattern. Of course, a change in the evaluation category of other basic classification items may be determined as another action pattern.

  The reference point setting unit 27 is a functional unit that sets a guidance reference point according to a predetermined condition (guidance reference point setting condition). The guidance reference point is a point serving as a reference for determining a point that can be a destination candidate included in the guidance information G. In such a guidance reference point, a destination (including a transit point) designated by the target user, a self-location indicated in the self-location information, and the like can be set. In the present embodiment, the guidance reference point setting condition is a condition regarding whether or not the destination is designated by the target user. When the destination is designated by the target user, the reference point setting unit 27 automatically sets the designated destination as the guidance reference point. When the destination is not specified, the reference point setting unit 27 may automatically set its own position as the guidance reference point. The reference point setting unit 27 may set its own position as the guidance reference point even when the destination is specified, particularly when the target user specifies it. Moreover, you may set another point as a guidance reference point.

  The positional relationship pattern determination unit 30 is a functional unit that determines a target positional relationship pattern that matches the positional relationship between the home position of the target user and the guidance reference point from a plurality of predetermined positional relationship patterns. In this embodiment, the positional relationship pattern determination unit 30 starts from the guidance reference point based on the guidance reference point and the home data K of the target user when the guidance reference point is set by the reference point setting unit 27. The distance to the home position and the relative orientation between the guidance reference point and the home position are calculated. Specifically, the positional relationship pattern determination unit 30 uses the position coordinates of the guidance reference point set by the reference point setting unit 27 and the home address (home data K) of the target user stored in the user database 44. The distance of the shortest route connecting the two positions is calculated from the identified position coordinates of the home, and the relative orientation of the target user's home around the guidance reference point is calculated. Then, using these calculated distances and relative orientations, the positional relationship pattern determination unit 30 refers to a plurality of predetermined positional relationship patterns (hereinafter referred to as reference positional relationship patterns) and determines the position of the target user from among them. A positional relationship pattern that matches the relationship is alternatively selected and determined as a target positional relationship pattern. The positional relationship pattern determination unit 30 is configured to calculate a linear distance between the position coordinates instead of calculating the distance of the shortest route connecting the position coordinates of the guidance reference point and the position coordinates of the target user's home. May be. Further, the positional relationship pattern determination unit 30 may be configured to calculate the relative orientation of the guidance reference point centered on the home position instead of the relative orientation of the home position centered on the guidance reference point.

  Here, the positional relationship pattern is defined by classifying the positional relationship between two points into a plurality based on one or more indices. The positional relationship pattern in the present embodiment includes at least one of a relative orientation between a point that can be a home position and a point that can be a guidance reference point, and a distance between the point that can be the home position and a point that can be a guidance reference point. Based on multiple categories. In the following, description will be given using an example in which the positional relationship patterns are classified into a plurality based on both the relative azimuth and the distance. In the present embodiment, the plurality of positional relationship patterns are divided into a plurality of relative orientations between a point that can be a home position and a point that can be a guidance reference point, a point that can be a home position, and a guidance reference. It is defined on the basis of a combination with a distance classification formed by dividing a distance from a possible point into a plurality of points. Specifically, the direction division is divided into a plurality of directions around a point that can be a home position and a point that can be a guide reference point. In this embodiment, as shown in FIG. 13, the direction division is divided into 8 directions (east, west, south, north, southeast, southwest, northeast, northwest) with the guide reference point as the center. In addition, the distance division is divided into a plurality of stages by comparing the distance from one of the points that can be the home position and the guide reference point with a predetermined reference distance X. . In this embodiment, the distance division is divided into three stages of “far”, “standard”, and “close”. More specifically, for example, a predetermined straight line reference distance X (between the two points is defined as a reference point that is one of two points of a home position and a guide reference point. A distance shorter than the connecting straight line distance) is set, the range from the reference point to the reference distance X + Y (Y is an arbitrary value smaller than X) and the reference distance X−Y is “standard”, from the reference point A range less than the reference distance X−Y is classified as “close”, and a range greater than the reference distance X + Y from the reference point is classified as “far”. A single positional relationship pattern is formed by combining these azimuth sections and distance sections.

  Examples of the positional relationship pattern include “Bβ”, “Aβ”, “Cdα”, “Cγ”, “Cdγ”, and the like as shown in FIG. More specifically, in the present embodiment, as shown in FIG. 13, A to D are assigned to “north”, “east”, “south”, and “west” of the orientation sections, respectively. In addition, Ab, Ad, Cb, and Cd are assigned to “northeast”, “northwest”, “southeast”, and “southwest”, respectively. Then, α, β, and γ are assigned to “distance”, “standard”, and “near”, which are distance categories, respectively. A storage device such as a user database 44 is configured so that these sections are combined to form respective positional relationship patterns, and a total of 512 patterns are defined in advance as reference positional relationship patterns and can be referred to by other functional units. (Stored). Stored in a storage device. As shown in FIG. 12, the positional relationship pattern of the user P1 whose relative orientation between the home position and the guidance reference point is “east” and the distance between the home position and the guidance reference point is “standard” is “ Bβ ". In addition, the positional relationship pattern of the user P5 whose relative orientation between the home position and the guidance reference point is “southeast” and whose distance is “close” is “Cbγ”.

  The positional relationship pattern determination unit 30 also has a function of determining a different positional relationship pattern different from the target positional relationship pattern from a plurality of predetermined positional relationship patterns. In the present embodiment, when determining the target positional relationship pattern, the positional relationship pattern determination unit 30 further refers to the reference positional relationship pattern, and combines the distance section and the azimuth section that define the target positional relationship pattern from among them. On the other hand, a positional relationship pattern in which either one of the distance segment and the azimuth segment differs is alternatively selected and determined as another positional relationship pattern. At this time, the positional relationship pattern determination unit 30 selects a positional relationship pattern that is relatively close to the target positional relationship pattern as another positional relationship pattern, or selects a positional relationship pattern that is significantly different from the target positional relationship pattern. Can be set as appropriate.

  As an example of the different positional relationship pattern, when the target positional relationship pattern is “Bβ”, “Bα”, “Bγ” having different azimuth divisions and different distance divisions as distance candidate pattern candidates, There are “Aβ”, “Abβ”, “Cbβ”, “Cβ”, “Cdβ”, “Dβ”, and “Adβ” that have the same section but different orientation sections. The positional relationship pattern determination unit 30 alternatively selects a positional relationship pattern from these candidates and determines it as another positional relationship pattern. For example, if you want to propose the second recommended point as an auxiliary point for the first recommended point, the second extraction condition is relatively similar to the target position relationship pattern in order to make the second extraction condition as close as possible to the first extraction condition. Select another positional relationship pattern. For example, “Cbβ” or “Abβ” having the same distance division and the adjacent azimuth division as the target positional relationship pattern “Bβ” is selected (see FIG. 13). In addition, a highly recommended point due to the positional relationship between the home position and the guidance reference point, such as a facility with a low level of recognition that only a person living in a different area from the target user knows the second recommended point. In order to make a proposal, in order to make the second extraction condition significantly different from the first extraction condition, another positional relationship pattern having a large difference from the target positional relationship pattern is selected. For example, the target positional relationship pattern “Bβ” and “Dβ” whose azimuth classification is the opposite are selected. In the above-described example, the positional relationship pattern with different azimuth segments is selected and determined as the different positional relationship pattern with respect to the combination of the distance segment and the azimuth segment that define the target positional relationship pattern. However, the configuration may be such that a positional relationship pattern in which only the distance classification is different from the target positional relationship pattern is selected and determined. Moreover, the structure which selects and determines the positional relationship pattern in which both the direction division and the distance division differ may be sufficient.

  Further, when the reference reference point setting unit 27 sets the guidance reference point, the positional relationship pattern determination unit 30 extracts the movement history data T of another user (from the second to be described later). Similarly to determining the target positional relationship pattern (when performing the fourth filtering), each positional relationship pattern that matches the positional relationship between the home position of another user and the guidance reference point is determined.

  The data extraction unit 28 is a functional unit that extracts the movement history data T including the guidance reference point from the movement history database 43 based on a predetermined extraction condition. The data extraction unit 28 includes a first data extraction unit 38 and a second data extraction unit. 39 and a third data extraction unit 40. The first data extraction unit 38 extracts the movement history data T including the guidance reference point from the movement history database 43 based on a first extraction condition that is associated with the target behavior pattern. In the present embodiment, the first extraction condition includes a further condition that the positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern. Accordingly, the first data extraction unit 38 is the movement history data T associated with the target behavior pattern and including the guidance reference point from the movement history database 43, and further the positional relationship between the home position and the guidance reference point. The movement history data T of the user whose pattern is the same as the target positional relationship pattern is extracted.

  The second data extraction unit 39 extracts the movement history data T including the guidance reference point from the movement history database 43 based on the second extraction condition that is different from the first extraction condition. In the present embodiment, the second extraction condition is that the positional relationship pattern between the home position and the guidance reference point is a different positional relationship pattern different from the target positional relationship pattern, and the movement history data T is a further condition. It includes being associated with the target behavior pattern. Therefore, the second data extraction unit 39 is the movement history data T associated with the target action pattern and including the guidance reference point from the movement history database 43, and further the positional relationship between the home position and the guidance reference point. The movement history data T of the user whose pattern is the same as the different positional relationship pattern is extracted.

  The third data extraction unit 40 extracts the movement history data T including the guidance reference point from the movement history database 43 based on a third extraction condition different from the first extraction condition and the second extraction condition. In the present embodiment, the third extraction condition is that the movement history data T is associated with another action pattern different from the target action pattern, and as a further condition, the positional relationship between the home position and the guidance reference point This includes that the pattern is the same as the target positional relationship pattern. Therefore, the third data extraction unit 40 is the movement history data T that is associated with the different action pattern and includes the guidance reference point from the movement history database 43, and further the positional relationship between the home position and the guidance reference point. The movement history data T of the user whose pattern is the same as the target positional relationship pattern is extracted. These first to third extraction conditions may be set for each operation by the user, may be set in advance by the user at the initial setting such as purchase, or may be set in advance at the time of shipment. Also good.

  Specifically, the data extraction unit 28 extracts data including information on the set guide reference point from the movement history data T stored in the movement history database 43. That is, the data extraction unit 28 performs the first filtering on all the movement history data T based on whether or not the guidance reference point currently set for the target user is included. In addition, the data extraction unit 28 (first data extraction unit 38) extracts data associated with the target behavior pattern (the behavior pattern that matches the behavior tendency of the target user) in the movement history data T after the first filtering. In addition to extraction, the movement history data T of the user whose positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern is extracted. That is, the data extraction unit 28 (first data extraction unit 38) is another user who shows the same behavior tendency as the target user for the movement history data T after the first filtering, and guides the home position and guidance. The second filtering is performed based on whether or not the movement history data T of the user whose positional relationship with the reference point is close to the target user.

  Further, the data extraction unit 28 (second data extraction unit 39) extracts data associated with the target behavior pattern from the movement history data T after the first filtering, and calculates the home position and the guidance reference point. The movement history data T of the user whose position relation pattern is the same as another position relation pattern is extracted. That is, the data extraction unit 28 (second data extraction unit 39) is another user who shows the same behavior tendency as the target user for the movement history data T after the first filtering, and guides the home position and guidance. The third filtering is performed based on whether or not the movement history data T of the user whose positional relationship with the reference point is different (not close) to the target user.

  Furthermore, the data extraction unit 28 (third data extraction unit 40) extracts data associated with another behavior pattern (behavior pattern different from the behavior tendency of the target user) from the movement history data T after the first filtering. In addition to extraction, the movement history data T of the user whose positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern is extracted. That is, the data extraction unit 28 (third data extraction unit 40) is another user who exhibits a behavior tendency different from the target user for the movement history data T after the first filtering, and guides the home position and guidance. The fourth filtering is performed based on whether or not the movement history data T of the user whose positional relationship with the reference point is close to the target user.

  In each of the second filtering to the fourth filtering, two-stage filtering processing is performed, but the order of the two-stage filtering processing is not particularly limited. For example, in the second filtering, the data previously associated with the target action pattern is extracted from the movement history data T after the first filtering, and then the positional relationship pattern between the home position and the guidance reference point is the target. The configuration may be such that user movement history data T that is the same as the positional relationship pattern is extracted, may be in the reverse order, or may be a configuration in which these are executed simultaneously.

  Further, the order of the first filtering to the fourth filtering is not particularly limited. Contrary to the above description, the first filtering based on the guidance reference point may be performed after the second to fourth filtering based on the action pattern and the position pattern. However, in this case, it is necessary to perform the first filtering on the movement history data T after the second filtering and after the fourth filtering, respectively. From the viewpoint of avoiding the redundancy of such filtering processing, it is preferable that the first filtering is performed collectively as described above.

  In any case, the movement history data T finally extracted through the first and second filtering is another user who matches the behavior tendency of the target user, and includes a home position, a guidance reference point, When the user who is close to the target user visits the guidance reference point in the past, information on other points visited is included. The movement history data T finally extracted through the first and third filtering is another user who matches the behavior tendency of the target user, and the positional relationship between the home position and the guidance reference point When the user who is in a different relationship with the target user (not in a close relationship) has visited the guidance reference point in the past, information on other points visited is included. In addition, the movement history data T finally extracted through the first and fourth filterings is another user having a behavior tendency different from that of the target user, and is the position of the home position and the guidance reference point. When a user whose relationship is close to that of the target user has visited the guidance reference point in the past, information on other points visited is included.

  The point-to-point correlation (referred to as “post-processing point-to-point correlation”) obtained based on the movement history data T after extraction processing of data associated with the target behavior pattern in the second filtering is all users. This may be different from the correlation between points based on the data (see FIG. 9). Further, the degree of correlation between post-processing points may be different between users having different target behavior patterns. For example, FIG. 10 shows the degree of correlation between post-processing points for a user with a relatively short moving distance between visited points. FIG. 11 shows the degree of correlation between post-processing points for a user who has a relatively long moving distance between visited points. It can be clearly understood that they exhibit completely different forms.

  For example, when attention is paid to the point D9, when all user data is considered, the degree of correlation with the point D1 is the highest (see FIG. 9). On the other hand, when filtering is performed for a user with a relatively short moving distance, the degree of correlation with the point D1 is almost lost and the degree of correlation with the point D8 is the highest (see FIG. 10). In addition, when filtering is performed for a user with a relatively long moving distance, the degree of correlation with the point D1 is low and the degree of correlation with the point D2 is the highest (see FIG. 11). Since the thickness of each line in FIGS. 9 to 11 conceptually shown is relative, a new line segment may appear as a result of the relative degree of correlation being increased by filtering. As described above, in the present embodiment, the degree of correlation between post-processing points that is highly likely to match the behavior tendency regarding the movement mode between the visited points is determined for each target user according to the target behavior pattern. After the extraction process of the data associated with the target behavior pattern in the third filtering and between the points grasped based on the movement history data T after the extraction process of the data associated with another behavior pattern in the fourth filtering The degree of correlation can be considered similarly.

  The recommended point presentation unit 29 is a functional unit that presents recommended points, which are points recommended as destination candidates, to the target user. The recommended spot presentation unit 29 determines a recommended spot based on the movement history data T extracted by the data extraction unit 28 and presents it to the target user. The recommended spot presentation unit 29 presents three types of recommended spots (first recommended spot / second recommended spot / third recommended spot) having different characteristics. Specifically, the recommended spot presentation unit 29 is included in the movement history data T extracted by the first data extraction unit 38, that is, the movement history data T finally extracted through the first and second filtering. A point having a relatively high movement frequency with the guidance reference point is determined as the first recommended point. In other words, the recommended spot presenting unit 29 determines a spot having a relatively high degree of correlation between post-processing spots based on the target behavior pattern with respect to the guidance reference spot as the first recommended spot.

  Further, the recommended point presentation unit 29 is connected to the guidance reference point included in the movement history data T extracted by the second data extraction unit 39, that is, the movement history data T extracted by the first and third filtering. A point having a relatively high movement frequency is determined as the second recommended point. In other words, the recommended point presentation unit 29 determines a point having a relatively high degree of correlation between post-processing points based on the target behavior pattern with the guide reference point as a reference as the second recommended point.

  Further, the recommended point presentation unit 29 is connected to the guidance reference point included in the movement history data T extracted by the third data extraction unit 40, that is, the movement history data T extracted by the first and fourth filtering. A point having a relatively high movement frequency is determined as the third recommended point. In other words, the recommended point presenting unit 29 determines a point having a relatively high degree of correlation between post-processing points based on another behavior pattern with the guidance reference point as a reference as the third recommended point.

  Then, the recommended point presentation unit 29 presents each of the determined recommended points as a candidate for a destination next to the currently set destination. That is, the recommended point presentation unit 29 first recommends a point having a relatively high movement frequency with respect to the guidance reference point from the movement history data T extracted by the data extraction unit 28 based on the first extraction condition. A point that is presented to the target user as a point and a point having a relatively high movement frequency with respect to the guidance reference point among the movement history data T extracted based on the second extraction condition by the data extraction unit 28 is second recommended. Present as a point with the first recommended point. Further, the recommended spot presenting unit 29 recommends a point having a relatively high movement frequency with respect to the guidance reference point from the movement history data T extracted by the data extracting unit 28 based on the third extraction condition. The point is presented together with the first recommended point and the second recommended point. In this embodiment, when calculating the movement frequency between the guidance reference point, both the movement from the guidance reference point to another visit point and the movement from another visit point to the guidance reference point are Be counted.

  As described above, the movement history data T extracted by the first and second filtering is another user that matches the behavior tendency of the target user, and the positional relationship between the home position and the guidance reference point is the target user. When the user who has a close relationship with visits the guidance reference point in the past, information on other points visited is included. In other words, the recommended point presenting unit 29, as a first recommended point, visited in addition to a visitor who visited a guidance reference point in the past when a user who also resembles the target user and the area where the behavior tendency lives (home position) is also visited. Present the point. Therefore, a point that is likely to be visited by the target user due to the positional relationship between the home position such as a point close to the target user's home position and the guidance reference point is a point that the target user is likely to like. It can be presented as a recommended spot.

  The movement history data T extracted by the first and third filtering is another user that matches the behavior tendency of the target user, and the positional relationship between the home position and the guidance reference point is close to the target user. It includes information on other points visited when a user who is not in the past visited the guidance reference point in the past. That is, the recommended point presenting unit 29, as the second recommended point, has a behavior tendency similar to that of the target user, but a user having a different living area (home position) or the like has visited the guidance reference point in the past. Present other points visited in time. Therefore, although it is a point that the target user is likely to like, the target user is caused by the positional relationship between the home position and the guidance reference point, such as a generally unknown point in a region away from the target user's home position. A point that is unlikely to be visited can be presented as the second recommended point.

  In addition, the movement history data T extracted by the first and fourth filtering is another user having a behavior tendency different from that of the target user, and the positional relationship between the home position and the guidance reference point is close to the target user. The information includes information on other points visited when a related user visited the guidance reference point in the past. In other words, the recommended point presentation unit 29 has a behavior tendency different from that of the target user as the third recommended point, but when a user with a similar living area (home position) visits the guidance reference point in the past. Present other points visited. Therefore, it is a point where the target user is highly likely to visit due to the positional relationship between the home position such as a point close to the target user's home position and the guidance reference point, but the unexpectedness is different from the target user's preference. A point can be presented as a third recommended point.

  When the recommended point presentation unit 29 determines each of the first recommended point, the second recommended point, and the third recommended point, the recommended point is determined as the next destination of the currently set destination. Present as a candidate. For example, the guidance information G to that effect is displayed in a guidance frame set in the display area of the display input device 36 or is uttered from the voice output device 37. In this embodiment, the most recommended point as the next destination candidate of the currently set destination is presented as the first recommended point, and the next recommended point is presented as the second recommended point. Then, the next recommended spot is presented as the third recommended spot. Accordingly, in the present embodiment, the first recommended spot is highlighted more than the second recommended spot, and the second recommended spot is highlighted more than the third recommended spot. FIG. 14 shows a display example of these recommended points. In the example of FIG. 14, the currently set destination is the point (D9), the first recommended point (the most likely candidate for the next destination) is the point (D2), and the second recommended point is the point ( D11), and the third recommended point is the point (D8). In the illustrated example, the first recommended spot is displayed more prominently in a relatively wide guide frame area in the form of text data with image data, and the second recommended spot is the other recommended spot in the guide frame area. In the recommended spot display area, it is highlighted above the third recommended spot by being displayed above the third recommended spot.

2. Navigation Process Procedure A navigation process procedure (navigation method) executed in the navigation system 1 according to the present embodiment will be described. The procedure of the navigation process described below is executed by hardware, software (program), or both constituting each functional unit of the navigation system 1. When each functional unit is configured by a program, the arithmetic processing device included in the navigation system 1 operates as a computer (for realizing each function) that executes the program that configures each functional unit described above.

FIG. 15 is a flowchart showing a procedure for generating the action history database 42 in each in-vehicle terminal device 2. In the action history database generation process, first, the date of the day and the departure time of the home (or the previous night's accommodation) are recorded (step # 01). It is determined whether or not the own position indicated in the own position information is located outside the living area for the user (# 02). The range of the living area is set in advance for each user based on the home position.
If it is determined that it is out of the living area (# 02: Yes), then the presence of a visit point (a point set as a destination or a point determined to have stayed for a predetermined time or more) is confirmed (# 03). If it is determined that there is a visit point (# 03: Yes), the point ID, arrival time, and stay time for the visit point are recorded (# 04). The processes of steps # 03 and # 04 are repeatedly executed until the user arrives at his / her home (or staying place of the day). When the return is determined (# 05: Yes), the return time is recorded (# 06). As described above, the action history data A of the day is generated, and the action history database 42 is generated (updated) based on the action history data A (# 07).

  FIG. 16 is a flowchart showing the procedure of the generation process of the movement history database 63 in the management server 5. In the movement history database generation process, first, action history data A for each user is collected from each in-vehicle terminal device 2 (# 11). Based on the collected action history data A, an action history database 62 including action history data A of all users is generated (# 12). At this time, the action history data A is stored in a state associated with the user ID. The behavior history data A for each user is extracted based on the user ID (# 13), and the behavior pattern for each user is determined by applying a statistical method to these (# 14). Further, the movement history data T is extracted from the action history data A (# 15). The action pattern determination and the movement history data T extraction may be performed in the reverse order. Based on the extracted movement history data T, a movement history database 63 including movement history data T for all users is generated (# 16). At this time, the movement history data T is stored in a state associated with the user ID and each behavior pattern. Then, upon receiving a distribution request from each in-vehicle terminal device 2 (# 17: Yes), the movement history data T is distributed (# 18).

  FIG. 17 is a flowchart illustrating a procedure of recommended spot presentation processing in each in-vehicle terminal device 2. In the recommended point presentation process, first, an action pattern (target action pattern) that matches the action tendency of the target user is determined (# 21), and a guidance reference point is set (# 22). Then, based on the guidance reference point set in step # 22 and the home data K of the target user, the target position relationship pattern is determined (# 23). Further, an action pattern (different action pattern) different from the target action pattern determined in step # 21 is further determined (# 24), and a position relation pattern (separate from the target position relation pattern determined in step # 23) is determined. The positional relationship pattern) is further determined (# 25). Note that step # 21, step # 22, and step # 24 may be executed in an order different from the above. Step # 23 and step # 25 may be executed in an order different from the above order as long as it is after step # 22. Next, data including information on the guidance reference point is extracted from all the movement history data T (# 26: first filtering). Subsequently, from the movement history data T extracted by the first filtering in step # 26, the target behavior pattern determined in step # 21 based on the first extraction condition (the behavior pattern that matches the behavior tendency of the target user). (# 27) and data of users having a positional relationship pattern that matches the target positional relationship pattern determined in step # 23 from the movement history data T extracted in step # 27 Are extracted (# 28: second filtering). It should be noted that step # 28 includes a process for determining the positional relationship pattern of users other than the target user. The movement frequency with respect to the guidance reference point included in the movement history data T after the filtering process (first and second filtering) of step # 26, step # 27, and step # 28 is relatively A high spot is determined as the first recommended spot (# 29). In addition, this step # 27, step # 28, and step # 29 are 1st recommendation spot determination processes.

  Next, data associated with the target action pattern determined in step # 21 is extracted from the movement history data T extracted by the first filtering in step # 26 based on the second extraction condition (# 30). ) And user data having another target positional relationship pattern determined in step # 25 are extracted from the movement history data T extracted in step # 30 (# 31: third filtering). In the extraction process at step # 31, the positional relationship pattern of the other user determined at step # 28 is used. And the movement frequency between the guidance reference points included in the movement history data T after the filtering process (first and third filtering) of Step # 26, Step # 30, and Step # 31 is relatively The high spot is determined as the second recommended spot (# 32). In addition, this step # 30, step # 31, and step # 32 are 2nd recommendation spot determination processes. Further, since step # 30 is the same process as step # 27, step # 30 is omitted, and in step # 31, another object determined in step # 25 from the movement history data T extracted in step # 27. A configuration may be adopted in which user data having a positional relationship pattern is extracted.

  Subsequently, from the movement history data T extracted by the first filtering in step # 26, data associated with the different target action pattern determined in step # 24 is extracted based on the third extraction condition (# 33), user data having the target positional relationship pattern determined in step # 23 is extracted from the movement history data T extracted in step # 33 (# 34: fourth filtering). In the extraction process at step # 34, the positional relationship pattern of the other user determined at step # 28 is used. And the movement frequency between the guidance reference points included in the movement history data T after the filtering process (first and fourth filtering) of Step # 26, Step # 33, and Step # 34 is relatively The high spot is determined as the third recommended spot (# 35). Note that Step # 33, Step # 34, and Step # 35 are the third recommended spot determination process. These first recommended point, second recommended point, and third recommended point are presented to the target user as candidates for the next destination (# 36). Note that the first recommended spot determination process, the second recommended spot determination process, and the third recommended spot determination process may be executed in any order.

3. Other Embodiments Finally, other embodiments of the navigation system according to the present invention will be described. Note that the configurations disclosed in the following embodiments can be applied in combination with the configurations disclosed in other embodiments as long as no contradiction arises.

(1) In the above embodiment, an example has been described in which the reference behavior pattern determination unit 26 determines one specific behavior pattern different from the target behavior pattern as another behavior pattern. However, the embodiment of the present invention is not limited to this. Another behavior pattern may be defined as a logical sum (set) of a plurality of behavior patterns other than the target behavior pattern. For example, it may be defined as a logical sum of all action patterns other than the target action pattern.

(2) In the above embodiment, the behavior pattern in which the reference behavior pattern determination unit 26 differs in the evaluation category of one specific classification basic item with respect to the combination of the evaluation categories of each classification basic item that defines the target behavior pattern The example which determines as another action pattern was demonstrated. However, the embodiment of the present invention is not limited to this. The different behavior pattern may be determined as a behavior pattern in which the evaluation classification of two or more specific classification basic items is different with respect to the combination of the evaluation classification of each classification basic item defining the target behavior pattern. As the number of classification basic items with different evaluation categories increases, the degree of deviation from the target behavior pattern increases, so the point preferred by the user who shows a behavior tendency that is significantly different from the behavior tendency of the target user is the second recommended point Increased chance of being presented. In this case, a more unexpected point can be proposed to each user as the second recommended point.

(3) In the above embodiment, the recommended point presentation unit 29 determines the one point having the highest movement frequency from the guidance reference point as the first recommended point, the second recommended point, and the third recommended point. The example presented to the target user has been described. However, the embodiment of the present invention is not limited to this. The recommended point presentation unit 29 may present a plurality of points including a point having the highest movement frequency with respect to the guidance reference point as the first recommended point, the second recommended point, and the third recommended point. One point having the highest movement frequency with the guidance reference point is set as the first recommended point, and a plurality of points including the point with the highest movement frequency with the guidance reference point are set as the second recommendation point and the third recommendation point. It may be presented as a recommended spot. Conversely, a plurality of points including a point having the highest movement frequency with respect to the guidance reference point are set as the first recommended point, and one point having the highest movement frequency with respect to the guidance reference point is set as the second recommended point or the second recommendation point. It may be presented as three recommended points.

(4) In the above embodiment, the recommended point presentation unit 29 has a relatively high movement frequency between the guide reference points without asking the moving direction (direction) between the two points including the guide reference points. An example in which a spot is determined as a recommended spot and presented to the target user has been described. However, the embodiment of the present invention is not limited to this. For example, the recommended point presenting unit 29 may present a point having a relatively high frequency as the destination from the guide reference point as the recommended point based on the moving direction (direction) between the two points.

(5) In the above embodiment, when extracting the movement history data T from the action history data A, the example in which only the movement between the directly moved points is extracted has been described. However, the embodiment of the present invention is not limited to this. For example, indirect movement between points, excluding the way of a visit point on the way, may be extracted (D1 → D2, D1 → D12, etc. in the example of FIG. 5). In this case, the movement history data T related to movement between indirect points may be weighted smaller than the movement history data T related to movement between direct points. That is, when counting the movement frequency between points, the movement between points may be counted with a value smaller than “1”.

(6) In the above embodiment, the recommended point presenting unit 29 uses the determined first recommended point, second recommended point, and third recommended point as candidates for the next destination next to the currently set destination. The example to be presented has been described. However, the embodiment of the present invention is not limited to this. For example, the recommended spot presenting unit 29 presents at least one of the first recommended spot, the second recommended spot, and the third recommended spot as a stopover candidate until the destination set at present is reached. You may do it.

(7) In the above embodiment, an example has been described in which each user's behavior pattern is classified based on the start time of specific behavior, such as departure time and lunch time. However, the embodiment of the present invention is not limited to this. For example, even if each user's action pattern is classified based on the return time, which is the end time of the entire day travel action, the end time of a specific event (for example, a meal, etc.) in the day action, etc. good.

(8) In the above embodiment, the three-level evaluation is performed for each of the travel distance between the visit points as the basic classification items, the stay time at the visit point, the departure time, and the lunch time, and the combination of the three evaluation categories. An example in which behavior patterns are classified based on the above has been described. However, the embodiment of the present invention is not limited to this. The number of basic classification items and the number of stages of evaluation categories can be arbitrary. The classification basic items may be fewer or more than the example in the above embodiment. Also regarding the number of stages of the evaluation category, it may be smaller or larger than the example in the above embodiment. Various items can be set as the contents of the basic classification items.

(9) In the above embodiment, when evaluating each basic classification item for behavior pattern classification, representative values having the estimated effective data remaining after deleting the estimated invalid data from all the data as the population An example of calculating and determining a behavior tendency based on the representative value has been described. However, the embodiment of the present invention is not limited to this. For example, it is possible to calculate a representative value using all data including data that may be irregular as a population, and determine a behavior tendency based on the representative value.

(10) In the above embodiment, an example in which the action history of the target user is reflected when the management server 5 generates (updates) the movement history database 63 has been described. However, the embodiment of the present invention is not limited to this. For example, the in-vehicle terminal device 2 corresponding to the target user may be configured not to transmit the action history data A to the management server 5. And the vehicle-mounted terminal device 2 is comprised so that a destination recommendation function may be implement | achieved using the movement history data T produced | generated based on the action history data A collected exclusively from the other vehicle-mounted terminal device 2. FIG. Also good. In this case, the movement history data T may be distributed from the management server 5 or may be stored (stored) in advance in the in-vehicle terminal device 2.

(11) In the above-described embodiment, the in-vehicle terminal device 2 includes the data extraction unit 28 and the recommended spot presentation unit 29, and the in-vehicle terminal apparatus 2 performs data extraction processing, recommended spot determination processing, and recommended spot presentation processing. An example to be described. However, the embodiment of the present invention is not limited to this. For example, the data extraction process and the recommended spot determination process among these may be executed on the management server 5 side. In this case, the management server 5 includes a data extraction unit and a recommended point determination unit, and the recommended points are determined by these functional units in the same manner as in the above embodiment. The determined recommended spot information is transmitted to the in-vehicle terminal device 2 of the target user, and the recommended spot is presented as a destination candidate.

(12) In the above embodiment, the example in which the aggregated movement history data T is directly distributed via the communication network 3 between the management server 5 and the in-vehicle terminal device 2 has been described. However, the embodiment of the present invention is not limited to this. The movement history data T may be distributed through the user's hand. For example, each user who has downloaded the movement history data T distributed via the communication network 3 using a personal computer at home or the like transfers the acquired movement history data T to the in-vehicle terminal device 2 using a medium such as a flash memory. May be. Alternatively, each user may transfer the movement history data T distributed in a state of being stored (stored) in a medium such as a flash memory or an optical disk to the in-vehicle terminal device 2.

(13) The above embodiment has been described assuming that the in-vehicle terminal device 2 constituting the navigation system 1 is a stationary navigation device fixed to the vehicle. However, the embodiment of the present invention is not limited to this. The vehicle-mounted terminal device 2 should just be mounted in the vehicle with the user at least, and does not necessarily need to be fixed to the vehicle. Examples of such a non-installation type navigation device include a PND (Portable Navigation Device), a multi-function mobile phone equipped with a predetermined navigation application and the like. In addition, when using these apparatuses, it does not necessarily need to be mounted on the vehicle. That is, the navigation system 1 according to the present embodiment may be configured by a mobile terminal device having a navigation function and a management server 5 provided to be able to communicate with a plurality of mobile terminal devices.

(14) In the above embodiment, when the positional relationship pattern determination unit 30 determines the positional relationship pattern of the target user and other users, the guidance reference point and the position coordinates of the home of each user (including the target user) The configuration is determined based on the above. However, the embodiment of the present invention is not limited to this. For example, instead of the location coordinates of each user's home, the administrative divisions and other areas such as prefectures and cities where each user's home is located are identified as the home surrounding area from the registered home phone number or postal code. The position relationship pattern may be determined using the representative point of the area around the home (for example, the position of an administrative agency such as a prefectural office or city hall or the center of gravity when the area is regarded as a plane) as the home position. In addition, since the area where the user's movement history points are gathered is highly likely to be within the user's living area, the movement history points are gathered if the home address or the like is not registered by the user. The configuration may be such that the positional relationship pattern is determined with the representative point (such as the average position) of the area as the home position.

(15) In the above embodiment, the positional relationship pattern includes a relative orientation between a point that can be a home position and a point that can be a guide reference point, and a point that can be the home position and a point that can be a guide reference point. It demonstrated using the example classified into multiple based on both of distance. However, the embodiment of the present invention is not limited to this. For example, classification based only on the relative orientation of a point that can be a home position and a point that can be a guide reference point, a classification based only on the distance between a point that can be a home position and a point that can be a guide reference point, etc. It is also possible to classify based on the indicators.

(16) In the above embodiment, the direction division is configured to be divided into eight directions (east, west, south, north, southeast, southwest, northeast, northwest) with the guide reference point as the center. However, the embodiment of the present invention is not limited to this. For example, the orientation division may be configured to be divided into four directions of east, west, north, and south, or may be configured to be divided into 32 directions. In the above-described embodiment, the distance division is divided into three stages of “far”, “standard”, and “close”. However, the two stages of “far” and “close” are used. It may be a configuration divided into four stages such as “far”, “slightly far”, “slightly close”, “close”, and the like. Moreover, the structure divided into five steps or more may be sufficient.

(17) In the above embodiment, the first to fourth filtering processes are performed, and the first recommended spot, the second recommended spot, and the third recommended spot are presented. However, the embodiment of the present invention is not limited to this. For example, a configuration in which only the first to third filtering processes are performed, that is, a configuration in which only the first recommended spot and the second recommended spot are presented may be used. Further, a configuration in which only the first and second filtering processes are performed, that is, a configuration in which only the first recommended spot is presented, or only the first, second, and fourth filtering processes are performed. In other words, only the first recommended point and the third recommended point may be presented.

(18) In the above embodiment, the data extraction unit 28 (second data extraction unit 39) is associated with the target action pattern in the movement history data T after the first filtering in the third filtering process. While extracting data, it was the structure which extracts the movement history data T of the user whose positional relationship pattern of a home position and a guidance reference point is the same as another positional relationship pattern. However, the embodiment of the present invention is not limited to this. For example, the data extraction unit 28 (second data extraction unit 39) extracts data associated with another behavior pattern from the movement history data T after the first filtering in the third filtering process, The configuration may be such that the movement history data T of the user whose position relation pattern between the position and the guidance reference point is the same as the other position relation pattern is extracted. Further, the third filtering is the same as that in the above embodiment, and the data extraction unit 28 (third data extraction unit 40) includes the movement history data T after the first filtering in the fourth filtering process. The configuration may be such that data associated with another behavior pattern is extracted, and movement history data T of a user whose positional relationship pattern between the home position and the guidance reference point is the same as the different positional relationship pattern is extracted.

(19) In the above embodiment, the data extraction unit 28 (first data extraction unit 38, third data extraction unit 40) specifies a user having a positional relationship pattern that matches the target positional relationship pattern, and the user's The movement history data T is extracted. However, the embodiment of the present invention is not limited to this. For example, when there is no user having a positional relationship pattern that matches the target positional relationship pattern, the data extraction unit 28 extracts the movement history data T of the user having the positional relationship pattern most similar to the target positional relationship pattern. You may comprise. Here, the positional relationship pattern most similar to the target positional relationship pattern is the case where the classification based on the relative orientation matches but the classification based on the distance does not match, or the classification based on the distance matches, but the classification based on the relative orientation Includes a case in which they are not matched and are classified into directions adjacent to each other. Specifically, if the target user is P3, the target positional relationship pattern is “Cdα” as shown in FIG. When there is no user having the positional relationship pattern “Cdα”, the classification based on the distance is different between “α” and “γ”, but the classification based on the relative orientation is common to “Cdγ”. P5 having a positional relationship pattern “” may be a user having a positional relationship pattern that matches the target positional relationship pattern, and the movement history data T of P5 may be extracted. This configuration is suitable when a positional relationship pattern that is significantly different from the target positional relationship pattern is selected as another positional relationship pattern.

(20) Regarding other configurations, the embodiments disclosed in this specification are merely examples in all respects, and the embodiments of the present invention are not limited thereto. In other words, configurations that are not described in the claims of the present application can be modified as appropriate without departing from the object of the present invention.

  The present invention can be used in a navigation system capable of executing a destination recommendation function.

1: Navigation system 26: Reference behavior pattern determination unit (target behavior pattern determination unit)
27: reference point setting unit 28: data extracting unit 29: recommended point presenting unit 30: positional relationship pattern determining unit 42: behavior history database 43: movement history database 44: user database 52: data collecting unit 54: history behavior pattern determining unit 55: Movement history database generation unit (database generation unit)
62: Action history database 63: Movement history database A: Action history data T: Movement history data G: Guidance information K: Home data

Claims (12)

A navigation system that provides guidance information related to points that can be candidates for a destination,
A movement history database that classifies and defines user behavior trends into a plurality of behavior patterns, stores movement history data in which movement histories between points of a plurality of users are recorded in association with any of the behavior patterns, and
A target behavior pattern determination unit that determines a target behavior pattern that matches a behavior tendency of a target user to be provided with the guidance information from the plurality of behavior patterns;
A reference point setting unit that sets a guide reference point according to a predetermined condition;
From the movement history database, based on each of the first extraction condition that is associated with the target behavior pattern and the second extraction condition that is different from the first extraction condition, the guidance criterion A data extraction unit for extracting the movement history data including a point;
Of the movement history data extracted based on the first extraction condition by the data extraction unit, a point having a relatively high movement frequency with the guide reference point is presented to the target user as a first recommended point. In addition, among the movement history data extracted based on the second extraction condition by the data extraction unit, a point having a relatively high movement frequency with the guidance reference point is set as the second recommended point. A navigation system comprising: a recommended point presentation unit for presenting with one recommended point. A user database storing home data which is information indicating the home position of each of the plurality of users;
A positional relationship pattern determination unit that determines a target positional relationship pattern that matches a positional relationship between the home position of the target user and the guidance reference point from a plurality of predetermined positional relationship patterns;
The first extraction condition includes as a further condition that the positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern,
The navigation system according to claim 1, wherein the second extraction condition is that the positional relationship pattern between the home position and the guidance reference point is a different positional relationship pattern different from the target positional relationship pattern.   The navigation system according to claim 2, wherein the second extraction condition includes a further condition that the movement history data is associated with the target behavior pattern.   The positional relationship pattern includes at least one of a relative orientation between a point that can be the home position and a point that can be the guidance reference point, and a distance between the point that can be the home position and the point that can be the guidance reference point. The navigation system according to claim 2 or 3, wherein the navigation system is classified into a plurality of categories. The plurality of positional relationship patterns are divided into a plurality of relative orientations of a point that can be the home position and a point that can be the guidance reference point, a point that can be the home position, and the guidance reference. Stipulated based on a combination with a distance class that is a number of distances to a point that can be a point,
5. The different positional relationship pattern is a positional relationship pattern in which either the distance segment or the azimuth segment is different from the combination of the distance segment and the azimuth segment that define the target positional relationship pattern. The navigation system described in. The data extraction unit extracts the movement history data including the guidance reference point based on a third extraction condition different from the first extraction condition and the second extraction condition,
The third extraction condition is that the movement history data is associated with another behavior pattern different from the target behavior pattern,
The recommended point presenting unit provides a third recommendation for a point having a relatively high movement frequency with respect to the guide reference point among the movement history data extracted based on the third extraction condition by the data extracting unit. The navigation system according to any one of claims 2 to 5, wherein the navigation system presents the first recommended spot and the second recommended spot as a spot.   The navigation system according to claim 6, wherein the third extraction condition includes a further condition that a positional relationship pattern between the home position and the guidance reference point is the same as the target positional relationship pattern. A multi-level evaluation is possible, and the basic items for classifying each behavior pattern are defined as classification basic items, and the categories representing the respective evaluation levels of the classification basic items are defined as evaluation categories,
The plurality of behavior patterns are respectively defined based on a plurality of evaluation categories for each of a plurality of basic classification items,
The navigation system according to claim 6 or 7, wherein the different behavior pattern is a behavior pattern in which an evaluation classification of one classification basic item is different from a combination of evaluation classification of each classification basic item defining the target behavior pattern. . The plurality of classification basic items include at least matters relating to the length of the moving distance between the destinations,
The said another action pattern is an action pattern from which the evaluation classification regarding the length of the movement distance between destinations differs with respect to the combination of the evaluation classification of each classification | category basic item which prescribe | regulates the said object action pattern. Navigation system. When the destination is designated by the target user, the reference point setting unit sets the designated destination as the guidance reference point,
The navigation system according to any one of claims 1 to 9, wherein the recommended spot presentation unit presents the first recommended spot and the second recommended spot as candidates for the next destination. A navigation method for providing guidance information related to a point that can be a destination candidate,
By classifying and defining user behavior trends into a plurality of behavior patterns, using a movement history database storing movement history data in which movement histories between points of a plurality of users are recorded in association with any of the behavior patterns,
A target behavior pattern determination step for determining a target behavior pattern that matches a behavior tendency of a target user to be provided with the guidance information from the plurality of behavior patterns;
A reference point setting step for setting a guide reference point according to a predetermined condition;
From the movement history database, based on each of the first extraction condition that is associated with the target behavior pattern and the second extraction condition that is different from the first extraction condition, the guidance criterion A data extraction step for extracting the movement history data including a point;
Of the movement history data extracted based on the first extraction condition by the data extraction step, a point having a relatively high movement frequency with the guidance reference point is presented to the target user as a first recommended point. In addition, among the movement history data extracted based on the second extraction condition by the data extraction step, a point having a relatively high movement frequency with the guide reference point is set as the second recommended point. A recommended point presentation step to be presented together with one recommended point;
Navigation method including. A navigation program that provides guidance information related to points that can be destination candidates,
A user's behavior tendency is classified and defined into a plurality of behavior patterns, referring to a movement history database storing movement history data in which movement histories between points of a plurality of users are recorded in association with any of the behavior patterns,
A target behavior pattern determination function for determining a target behavior pattern that matches a behavior tendency of a target user to be provided with the guidance information from the plurality of behavior patterns;
A reference point setting function for setting a guide reference point according to predetermined conditions;
From the movement history database, based on each of the first extraction condition that is associated with the target behavior pattern and the second extraction condition that is different from the first extraction condition, the guidance criterion A data extraction function for extracting the movement history data including points;
Of the movement history data extracted based on the first extraction condition by the data extraction function, a point having a relatively high movement frequency with the guide reference point is presented to the target user as a first recommended point. In addition, among the movement history data extracted based on the second extraction condition by the data extraction function, a point having a relatively high movement frequency with the guide reference point is set as the second recommended point. A recommended point presentation function to present with one recommended point;
A navigation program that enables computers to realize this.

Images