Travel route proposal system, travel route proposal method, and program

Abstract

This system provides a travel route suggestion system that, when a user sets a destination, can suggest intermediate stops that are likely to match the user's needs. [Solution] The travel route suggestion system 1 includes a user clustering unit 22 that classifies multiple users into multiple classes based on the users' behavior history information, a map information acquisition unit 14 that acquires map information of a target area including the destination set by the target user and the target user's current location, a suggested spot determination unit 18 that extracts candidate spots that can be passed through when traveling from the target user's current location to the destination based on the map information, and determines whether or not to make the candidate spots suggested spots to visit for the target user based on the behavior history information of users belonging to the target class to which the target user belongs, and a suggested spot information output unit 19 that outputs information about suggested spots.

Description

Technical Field

[0001] The present invention relates to a travel route proposal system, a travel route proposal method, and a program for proposing a travel route to a destination.

Background Art

[0002] In recent years, efforts have been actively made to provide access to a sustainable transport system that takes into account people in vulnerable positions among traffic participants. Toward this realization, research and development related to navigation technology has been focused on further improving traffic safety and convenience. For example, Patent Document 1 discloses a behavior plan providing method for providing a user with behavior plan information based on an estimated stay time at a planned visit destination of the user estimated from the user's behavior information and the user's preference information. Further, Patent Document 2 discloses an information processing apparatus that provides a next destination to a guided user based on the stay time at the destination of the guided user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] Incidentally, in navigation technology, it is conceivable to suggest intermediate stops that meet the user's needs in order to make travel to the destination set by the user more comfortable. However, the technologies described in the above-mentioned Patent Documents 1 and 2 only suggest destinations to the user, and therefore cannot suggest intermediate stops that meet the user's needs in accordance with the destination set by the user. Therefore, the object of this application is to suggest intermediate stops that meet the user's needs when the user sets a destination. This invention aims to provide a travel route suggestion system and method that can suggest intermediate stops that are likely to meet the user's needs when the user sets a destination, in order to solve the above-mentioned problems. Ultimately, this will contribute to the development of a sustainable transportation system. [Means for solving the problem]

[0005] As a first embodiment for achieving the above objective, a travel route proposal system is provided which includes: an action history information acquisition unit that acquires action history information showing the action history of multiple users; a user clustering unit that classifies the multiple users into multiple classes based on the action history information; a destination recognition unit that recognizes a destination set by a target user who is one of the multiple users; a user location recognition unit that recognizes the current location of the target user; a map information acquisition unit that acquires map information of a target area including the destination and the current location of the target user; a proposed spot determination unit that extracts candidate spots that can be passed through when moving from the current location of the target user to the destination based on the map information, and determines whether or not to make the candidate spots proposed spots that the target user should visit, based on the action history information of the users belonging to the target class to which the target user belongs; and a proposed spot information output unit that outputs information about the proposed spots.

[0006] The above-described travel route suggestion system may also be configured to include: a stop selection reception unit that accepts the user's selection of whether or not to stop at the suggested spot in response to the output of suggested spot information by the suggested spot information output unit; and a suggested route setting unit that, when the stop selection reception unit receives a selection indicating that a stop at the suggested spot is necessary, sets a suggested route from the user's current location to the destination via the suggested spot, based on the map information.

[0007] In the above-described travel route proposal system, the behavior history information includes proposal result data that associates information on the proposed spots with the results of the stop selection reception unit's selection of whether or not to stop at the proposed spots. The system also includes a route proposal model generation unit that generates a route proposal model for each of the multiple classes, which outputs a determination of whether or not to make a candidate spot a proposed spot in response to input information on a candidate spot, by performing machine learning using training data based on the proposal result data for the users belonging to that class. The proposed spot determination unit may be configured to determine whether or not to make a candidate spot a proposed spot using the route proposal model corresponding to the target class.

[0008] In the above-described travel route proposal system, the activity history information may include proposal result data that associates the information of the proposed spots, the user's travel status at the time the proposed spot information output unit outputs the information of the proposed spots, and the result of the stop selection acceptance unit's selection of whether or not to stop at the proposed spots. The route proposal model generation unit may be configured to generate the route proposal model that, in response to the input of the candidate spot information and the user's travel status, outputs a determination of whether or not to make the candidate spots the proposed spots.

[0009] In the above-described travel route proposal system, the route proposal model generation unit may be configured to generate the route proposal model by machine learning, by weighting candidate spots that have been selected as requiring a stop by the stop selection reception unit, giving them priority over candidate spots that have been selected as not requiring a stop by the stop selection reception unit.

[0010] In the above-described travel route suggestion system, the route suggestion model generation unit may be configured to retrain the route suggestion model corresponding to the class to which the user, who is the target of the added suggestion result data, belongs, using the newly added suggestion result data when new suggestion result data is added to the behavior history information.

[0011] The above-described travel route proposal system may be configured to include a desired arrival time recognition unit that recognizes the desired arrival time, which is the time when the target user will arrive at the destination, and a proposed spot determination unit that extracts as candidate spots spots that the target user can visit and that allow them to reach the destination by the desired arrival time.

[0012] As a second embodiment for achieving the above objective, a travel route proposal method performed by a computer is provided, which includes: an action history information acquisition step for acquiring action history information indicating the action history of multiple users; a user clustering step for classifying the multiple users into multiple classes based on the action history information; a destination recognition step for recognizing a destination set by a target user who is one of the multiple users; a user location recognition step for recognizing the current location of the target user; a map information acquisition step for acquiring map information of a target area including the destination and the current location of the target user; a proposed spot determination step for extracting candidate spots that can be passed through when traveling from the current location of the target user to the destination based on the map information, and determining whether or not to make the candidate spots proposed spots that the target user should visit based on the action history information of the users belonging to the target class to which the target user belongs; and a proposed spot information output step for outputting information about the proposed spots.

[0013] As a third embodiment for achieving the above objective, a program is provided that causes the computer to function as follows: an action history information acquisition unit that acquires action history information indicating the action history of multiple users; a user clustering unit that classifies the multiple users into multiple classes based on the action history information; a destination recognition unit that recognizes a destination set by a target user who is one of the multiple users; a user location recognition unit that recognizes the current location of the target user; a map information acquisition unit that acquires map information of a target area including the destination and the current location of the target user; a suggested spot determination unit that extracts candidate spots that can be passed through when moving from the current location of the target user to the destination based on the map information, and determines whether or not to make the candidate spots suggested spots to visit for the target user based on the action history information of the users belonging to the target class to which the target user belongs; and a suggested spot information output unit that outputs information about the suggested spots. [Effects of the Invention]

[0014] According to the above-described travel route proposal system, travel route proposal method, and program, when a user sets a destination, it is possible to propose transit spots that are likely to meet the needs of the user.

Brief Description of the Drawings

[0015] [Figure 1] FIG. 1 is a configuration diagram of the travel route proposal system according to the present embodiment. [Figure 2] FIG. 2 is an explanatory diagram of data stored in the storage device. [Figure 3] FIG. 3 is an explanatory diagram of clustering of multiple users. [Figure 4] FIG. 4 is an explanatory diagram of the generation process of the route proposal model. [Figure 5] FIG. 5 is a first flowchart of the process of proposing a stop at a proposed spot. [Figure 6] FIG. 6 is a second flowchart of the process of proposing a stop at a proposed spot. [Figure 7] FIG. 7 is an explanatory diagram of a series of processes from the determination of the proposed spot to the storage of the selection result of stopping at the proposed spot.

Modes for Carrying Out the Invention

[0016] [1. Configuration of the Travel Route Proposal System] Referring to FIG. 1, the configuration of the travel route proposal system 1 of the present embodiment will be described. The travel route proposal system 1 is a computer system connected to the communication network 100. The travel route proposal system 1 communicates with a communication terminal used by a user of the travel route proposal system 1 via the communication network 100, and provides the user with information on spots where a stop can be made during the journey to the destination.

[0017] In FIG. 1, as users of the travel route proposal system 1, user U1 who travels by riding in vehicle 50 and user U2 who travels on foot are illustrated. In FIG. 1, Ps is the current position of users U1 and U2, and Pd is the destination of users U1 and U2. The travel route proposal system 1 provides information on spots where one can stop for user U1 through communication with a communication terminal 51 (navigation device, DA (Display Audio), smartphone, tablet terminal, mobile phone, etc.) installed in vehicle 50 or brought in by user U1 into vehicle 50. Also, for user U2, the travel route proposal system 1 provides information on spots where one can stop through communication with a communication terminal 60 (smartphone, tablet terminal, mobile phone, etc.) carried by user U2.

[0018] Furthermore, the travel route proposal system 1 communicates with various servers such as traffic information server 110, spot information server 120, and vehicle support server 130 via communication network 100. The traffic information server 110 provides traffic information such as maps and road regulations. The spot information server 120 provides information on spots where one can stop (restaurants, service areas, drive-ins, tourist spots, etc.). The vehicle support server 130 receives and stores information on the travel status (travel location, travel time, travel speed, etc.) from service target vehicles such as vehicle 50.

[0019] The travel route suggestion system 1 includes a processor 10 (corresponding to the computer in this disclosure), a storage device 30, a communication unit 40, etc. The storage device 30 stores a control program 31 for the travel route suggestion system 1 (corresponding to the program in this disclosure), a machine learning platform program 32, a user database 33, and an activity history database 34. As shown in Figure 2, the user database 33 records registered user data 201 that associates user IDs (User01, User02, ...) issued individually to multiple users registered in the travel route suggestion system 1 with the class to which the user belongs (indicated as user class in Figure 2) and the registration information (communication address, home address, preferences, etc.) entered by each user at the time of registration. The class to which a user belongs is determined by a clustering process described later.

[0020] The activity history DB34 records, for each user, the proposal result data 200 and visit record data 202 shown in Figure 2. The proposal result data 200 and visit record data 202 correspond to the activity history information disclosed herein. The proposal result data 200 includes the date and time the proposal was made, the user's selection result regarding whether or not to visit the proposed spot (necessary or unnecessary), information about the proposed spot (proposal category, classification), and the user's travel status at the time the proposal was made (driving time, distance traveled, distance from home, number of stops on the day). The visit record data 202 records the number of visits to spots that the user has visited so far, categorized by genre. The visit record data 202 is used for user clustering and estimation of users' preferences regarding spots, as described later.

[0021] The processor 10 reads and executes the control program 31 for the travel route proposal system 1, and functions as a user location recognition unit 11, a destination recognition unit 12, a desired arrival time recognition unit 13, a map information acquisition unit 14, an activity history information acquisition unit 15, a user clustering unit 22, a route proposal model generation unit 16, a proposed spot determination unit 18, a proposed spot information output unit 19, a stopover selection acceptance unit 20, and a proposed route setting unit 21.

[0022] Here, the processing performed by the user location recognition unit 11 corresponds to the user location recognition step in the travel route proposal method of this disclosure, and the processing performed by the destination recognition unit 12 corresponds to the destination recognition step in the travel route proposal method of this disclosure. The processing performed by the map information acquisition unit 14 corresponds to the map information acquisition step in the travel route proposal method of this disclosure, and the processing performed by the activity history information acquisition unit 15 corresponds to the activity history information acquisition step in the travel route proposal method of this disclosure. The processing performed by the proposed spot determination unit 18 corresponds to the proposed spot determination step in the travel route proposal method of this disclosure, and the processing performed by the proposed spot information output unit 19 corresponds to the proposed spot information output step in the travel route proposal method of this disclosure. The processing performed by the user clustering unit 22 corresponds to the user clustering step in the travel route proposal method of this disclosure.

[0023] The following describes a scenario where user U1, who is driving vehicle 50 from their current location Ps to their destination Pd, is offered the option to stop at a spot along their route. The same applies when offering the option to stop at a spot to other users, such as user U2.

[0024] The user location recognition unit 11 communicates with the communication terminal 51 to acquire detection information of the user U1's current location (the current location of the vehicle 50) detected by a location sensor such as a GNSS (Global Navigation Satellite System) sensor installed in the communication terminal 51, and recognizes the current location of the user U1.

[0025] The destination recognition unit 12 communicates with the communication terminal 51 to recognize the destination Pd set by user U1 using the communication terminal 51. The desired arrival time recognition unit 13 communicates with the communication terminal 51 to recognize the desired arrival time at destination Pd set by user U1 using the communication terminal 51.

[0026] The map information acquisition unit 14 acquires map information of the target area, including the user U1's current location Ps and destination Pd, from the traffic information server 110 by accessing the traffic information server 110. The activity history information acquisition unit 15 acquires the vehicle 50's movement history information by communicating with the communication terminal 51 or by accessing the vehicle support server 130, and also recognizes the result of the acceptance of the stop selection acceptance unit 20 (described later) regarding whether or not to stop at the suggested spots, generates suggested result data 200, and records it in the activity history DB 34.

[0027] As shown in Figure 3, the user clustering unit 22 performs hierarchical clustering for multiple users using the number of visits to each category (Italian, Japanese, Chinese, etc.) recorded in the visit history data 202, which is aggregated for each user. The user clustering unit 22 then calculates a silhouette score and automatically sets the number of clusters that maximize the silhouette score. The class to which each user belongs, as classified by the user clustering unit 22 (Shopping Use and Weekend Leisure are examples in Figure 3), is recorded in the registered user data 201.

[0028] As shown in Figure 4, the route proposal model generation unit 16 classifies the proposal result data 200 accumulated for multiple users by referring to the class of each user recorded in the registered user data 201 (see Figure 2). Then, the route proposal model generation unit 16 individually generates a route proposal model 17 corresponding to each class using the proposal result data 200 classified by class.

[0029] Figure 4 illustrates the case of generating a route suggestion model corresponding to the "Shopping Use" class. The route suggestion model generation unit 16 processes the suggestion result data 200 of users belonging to the "Shopping Use" class into dummy variables by subdividing the suggested date and time (weekdays, holidays, days of the week, time slots) and the attributes of suggested spots (genre classification, purpose classification, etc.), and generates training data 210 in which the selection results are considered correct, using the information of suggested spots and travel status as examples.

[0030] The route suggestion model generation unit 16 then executes the machine learning platform program 32 and performs machine learning using the training data 210 to generate a route suggestion model 17 corresponding to the "shopping use" class, which, as shown in callout B1, outputs a determination of whether or not to suggest a candidate spot as a suggested spot in response to input information on candidate spots and the user's movement status. The route suggestion model generation unit 16 performs the same process for other classes as well, thereby individually generating a route suggestion model 17 corresponding to each class.

[0031] Here, when the route proposal model generation unit 16 generates the route proposal model 17 using machine learning, it may perform weighting to extract candidate spots that have been selected as necessary for a stop by the stop selection reception unit 20 as proposed spots, prioritizing them over candidate spots that have been selected as unnecessary for a stop by the stop selection reception unit 20.

[0032] The proposed spot determination unit 18 accesses the spot information server 120 to obtain information on spots located near user U1, thereby extracting candidate spots that user U1 can visit on their way to destination Pd. In this case, the proposed spot determination unit 18 extracts candidate steps on the condition that the user can reach destination Pd by the desired arrival time recognized by the desired arrival time recognition unit 13.

[0033] The proposed spot determination unit 18 inputs information on candidate spots and the user U1's movement status into the route proposal model 17, and, according to the output of the route proposal model 17's determination, decides whether or not to make the candidate spots proposed spots for user U1 to visit. The proposed spot information output unit 19 transmits information on proposed spots to the communication terminal 51 to propose a visit to the proposed spots. The visit selection reception unit 20 receives the selection result data on whether or not to visit the proposed spots transmitted from the communication terminal 51 and accepts the selection to visit the proposed spots.

[0034] When the stopover selection reception unit 20 receives a selection indicating that a stopover at a suggested spot is necessary, the suggested route setting unit 21 sets a suggested route from the user U1's current location (the vehicle 50's current location) to the destination Pd via the suggested spot, and transmits the suggested route information to the communication terminal 51. Upon receiving the suggested route information, the communication terminal 51 displays the suggested route information on its display and performs route guidance along the suggested route according to the user U1's instructions.

[0035] [2. Processing of proposals to visit suggested spots] Following the flowcharts shown in Figures 5 and 6, the procedure for suggesting stops at proposed spots, performed by the travel route suggestion system 1 for user U1 in the situation shown in Figure 1, will be explained with reference to the overview of the series of processes shown in Figure 7. In this case, user U1 corresponds to the target user of this disclosure.

[0036] When the travel route suggestion system 1 recognizes, through communication with the communication terminal 51, that user U1 has started traveling by vehicle 50, it executes a process to suggest stops at suggested spots according to the flowcharts in Figures 5 and 6. In step S1 of Figure 5, the travel route suggestion system 1 recognizes user U1's user ID through communication with the communication terminal 51 and identifies user U1.

[0037] In the following step S2, the destination recognition unit 12 communicates with the communication terminal 51 to determine whether or not a destination has been set for user U1. If a destination has been set, the process proceeds to step S3; otherwise, the process proceeds to step S20. In step S20, the destination recognition unit 12 communicates with the communication terminal 51 to determine whether or not user U1 has finished traveling by vehicle 50. If user U1 has finished traveling, the destination recognition unit 12 proceeds to step S18 in Figure 6; if user U1 is continuing to travel, the process proceeds to step S2.

[0038] In step S3, the user location recognition unit 11 recognizes the current location of user U1 (the current location of vehicle 50) through communication with the communication terminal 51. In the following step S4, the map information acquisition unit 14 acquires map information and road traffic information for the target area, including user U1's current location Ps and destination Pd, from the traffic information server 110. In the next step S5, the desired arrival time recognition unit 13 recognizes the desired arrival time at destination Pd set by user U1 through communication with the communication terminal 51.

[0039] In the next step S6, the proposed spot determination unit 18 refers to map information and traffic information to search for candidate spots that the user U1 can stop at when traveling from their current location Ps to destination Pd, provided that it is possible to reach destination Pd by the desired arrival time. Here, the proposed spot determination unit 18 searches for candidate spots that, depending on the type of spot (restaurant, service area, drive-in, convenience store, etc.), allow the user to reach destination Pd by the desired arrival time, assuming a typical stay time at each spot.

[0040] In the following step S7, the suggested spot determination unit 18 proceeds to S7 if candidate spots are extracted, and proceeds to step S18 in Figure 6 if no candidate spots are extracted, ending the stopover suggestion process. In step S8, as shown in the callout B2 in Figure 7, the suggested spot determination unit 18 recognizes the movement status of user U1 (movement status of vehicle 50) based on the behavior history data recognized by the behavior history information acquisition unit 15 and stored in the behavior history DB 34.

[0041] In the next step S9, the proposed spot determination unit 18 creates inference data 220, as shown in Figure 7, based on the candidate spot information and the user U1's travel status. The inference data 220 includes the class to which user U1 belongs (in this case, "Shopping Use"), candidate spot information (proposal category, classification), and user U1's travel status (time of day, driving time, distance traveled, number of stops on the day). In the following step S10, as shown in callout B3 in Figure 7, the proposed spot determination unit 18 inputs the inference data 220 into the route proposal model 17 corresponding to the "Shopping Use" class and obtains the output of the inference result (propose / do not propose).

[0042] If the collection of proposal result data 200 (see Figure 2) for user U1 is insufficient and a route proposal model 17 targeting user U1 has not yet been generated, the proposed spot determination unit 18 will, for example, determine candidate spots in genres that match user U1's preferences recorded in the registered user data 201 (see Figure 2) as proposed spots.

[0043] In the following step S11, the proposed spot determination unit 18 proceeds to step S12 if the output of the route proposal model 17 is "suggest", and proceeds to step S18 in Figure 6 if the output of the route proposal model 17 is "do not suggest". In step S12, the proposed spot information output unit transmits proposed spot information suggesting a stop at a proposed spot to the communication terminal 51, as shown in the callout B4 in Figure 7. Upon receiving the proposed spot information, the communication terminal 51 displays a stop suggestion screen 55 on its display, as shown in Figure 7.

[0044] The stop suggestion screen 55 displays that there is a restaurant nearby, and provides a Yes button 55a to select whether a stop is necessary and a No button 55b to select whether a stop is unnecessary. The communication terminal 51 transmits selection result information, indicating whether the user U1 pressed the Yes button 55a or the No button, or made a Yes or No selection via voice input, to the travel route suggestion system 1. In step S13 of Figure 6, when the stop selection reception unit 20 receives the selection result information transmitted from the communication terminal 51, it proceeds to step S14.

[0045] In step S14, the stopover selection reception unit 20 recognizes from the selection result information whether user U1 needs to stop at the suggested spot. If the selection result indicates that a stopover is necessary, the stopover selection reception unit 20 proceeds to step S15; if the selection result indicates that a stopover is not necessary, it proceeds to step S16. In the example in Figure 7, as shown in callout B5, user U1 selected that a stopover was not necessary and is heading to destination Pd (park) without stopping at the Italian restaurant 70.

[0046] In step S15, the proposed route setting unit 21 transmits information about the proposed route from the proposed spot to the destination Pd to the communication terminal 51. In the example in Figure 1, if restaurant 70 is the proposed spot, the proposed route setting unit 21 transmits information about the proposed route R1 from restaurant 70 to the destination Pd to the communication terminal 51. Also, if restaurant 71 is the proposed spot, the proposed route setting unit 21 transmits information about the proposed route R2 from restaurant 71 to the destination Pd to the communication terminal 51.

[0047] If, within a predetermined time from the time in step S12 when the suggested stop-visit information for the suggested spot is sent to the communication terminal 51, the selection result information sent from the communication terminal 51 is not received in step S13, the stop-visit selection reception unit 20 may determine that user U1 has selected that a stop-visit to the suggested spot is not necessary and proceed to step S16, or proceed to step S18.

[0048] In step S16, the behavior history information acquisition unit 15 generates the current proposal result data 200a, which includes the selection result of whether or not to stop at the proposed spots, as shown in callout B6 in Figure 7, and records it by adding it to the proposal result data 200 in the behavior history DB 34, as shown in callout B7. In the following step S17, the route proposal model generation unit 16 determines whether or not a predetermined number of proposal result data 200 have been added to the behavior history DB 34 for the class "Shopping Use," which is the target course for this time. If a predetermined number or more of proposal result data 200 have been added, the route proposal model generation unit 16 proceeds to step S30, and if a predetermined number or more of proposal result data 200 have not been added, it proceeds to step S18.

[0049] In step S20, the route suggestion model generation unit 16 updates the route suggestion model 17 corresponding to the "Shopping Use" class by performing the machine learning (retraining) described above, referring to Figure 4, including the added suggestion result data 200. By updating the route suggestion model 17 corresponding to the "Shopping Use" class in this way, reflecting the recent behavioral history of users belonging to the "Shopping Use" class, it is expected that the likelihood of determining suggested spots that match the needs of users belonging to the "Shopping Use" class will increase.

[0050] [3. Other Embodiments] In the above embodiment, the travel route proposal system of the present disclosure is configured by a travel route proposal system 1, which is a computer system (server) that communicates with a communication terminal 51 used in a vehicle 50 and a communication terminal 60 carried by a user U2 via a communication network 100. In other embodiments, the travel route proposal system of the present disclosure may be configured by communication terminals 51 and 60. Alternatively, a part of the travel proposal system of the present disclosure may be configured as a server function, and the remaining part may be configured by communication terminals 51 and 60. In this case, the travel route proposal system of the present disclosure is configured by the server and the communication terminal 51 or communication terminal 60. For example, the server may be provided with a route proposal model generation unit 16 and a proposed spot determination unit 18, which have a high computational load, while the other components may be provided by the communication terminals 51 and 60.

[0051] In the above embodiment, the route suggestion model generation unit 16 generated a route suggestion model 17 using training data 210 which includes information on suggested spots, the movement status of user U1, and the selection result by user U1 regarding whether or not to stop at the suggested spots, as shown in Figure 4. In another embodiment, training data may be used which does not include the movement status of user U1, but includes information on suggested spots and the selection result by user U1 regarding whether or not to stop at the suggested spots.

[0052] In the above embodiment, the travel route proposal system 1 includes a desired arrival time recognition unit 13, and the proposed spot determination unit 18 searches for candidate spots on the condition that the destination Pd is reached by the desired arrival time. In another embodiment, the desired arrival time recognition unit 13 may be omitted, and the proposed spot determination unit 18 may be configured to search for candidate spots without considering the desired arrival time.

[0053] In the above embodiment, the route suggestion model generation unit 16 retrained the route suggestion model 17 using the added suggestion result data, as shown in steps S17 and S30 of Figure 6. However, the configuration may also be configured so that the route suggestion model 17 is not retrained.

[0054] In the above embodiment, a route suggestion model generation unit 16 is provided, and the suggested spot determination unit 18 determines suggested spots using the route suggestion model 17 generated by the route suggestion model generation unit 16. Alternatively, suggested spots may be determined by other methods, such as determining spots that match the user's preferences and behavioral patterns estimated from the user's behavioral history, without using the route suggestion model 17 generated by machine learning.

[0055] Figure 1 is a schematic diagram showing the functional configuration of the travel path proposal system 1, divided by its main processing content, in order to facilitate understanding of the present invention. The functional configuration of the travel path proposal system 1 may be configured by other divisions. Furthermore, the processing of each component may be performed by one hardware unit or by multiple hardware units. Also, the processing of each component shown in Figures 5 and 6 may be performed by one program or by multiple programs.

[0056] [4. Configurations supported by the above embodiments] The above embodiment is a specific example of the following configuration.

[0057] (Configuration 1) A travel route proposal system comprising: an action history information acquisition unit that acquires action history information showing the action history of multiple users; a user clustering unit that classifies the multiple users into multiple classes based on the action history information; a destination recognition unit that recognizes a destination set by a target user who is one of the multiple users; a user location recognition unit that recognizes the current location of the target user; a map information acquisition unit that acquires map information of a target area including the destination and the current location of the target user; a proposed spot determination unit that extracts candidate spots that can be passed through when traveling from the current location of the target user to the destination based on the map information, and determines whether or not to make the candidate spots proposed spots that the target user should visit based on the action history information of the user belonging to the target class to which the target user belongs; and a proposed spot information output unit that outputs information about the proposed spots. According to the travel route suggestion system of Configuration 1, when a target user sets a destination, it can suggest intermediate stops that are likely to match the target user's needs, based on the behavioral history of users in the target class to which the target user belongs.

[0058] (Configuration 2) A travel route suggestion system according to Configuration 1, comprising: a stop selection reception unit that receives a selection from the target user regarding whether or not to stop at the suggested spot in response to the output of information on the suggested spot by the suggested spot information output unit; and a suggested route setting unit that, when the stop selection reception unit receives a selection that it is necessary to stop at the suggested spot, sets a suggested route from the user's current location to the destination via the suggested spot based on the map information. According to the travel route suggestion system of Configuration 2, a suggested route can be set to reach the destination via suggested spots, depending on whether the user selects whether or not it is necessary.

[0059] (Configuration 3) The travel route proposal system according to Configuration 2, wherein the behavior history information includes proposal result data that associates the information of the proposed spot with the result of the acceptance of the selection by the stop selection acceptance unit regarding whether or not to stop at the proposed spot, and for each of the multiple classes, a route proposal model generation unit generates a route proposal model for each of the multiple classes that outputs a determination of whether or not to make the candidate spot a proposed spot in response to the input of the candidate spot information, by performing machine learning using training data based on the proposal result data for the users belonging to the class, and the proposed spot determination unit determines whether or not to make the candidate spot a proposed spot using the route proposal model corresponding to the target class. According to the travel route suggestion system of Configuration 3, for each of the multiple classes, a route suggestion model corresponding to each class is generated by reflecting the results of users belonging to each class's selection of whether or not they need to stop at past suggested spots. Then, suggested spots for the target user can be determined using the route suggestion model corresponding to the target class to which the target user belongs.

[0060] (Configuration 4) The travel route proposal system according to Configuration 3, wherein the activity history information includes the information of the proposed spot, the travel status at the time of the proposal which is the user's travel status at the time the proposed spot information output unit outputs the information of the proposed spot, and the result of the selection by the stop selection reception unit regarding whether or not to stop at the proposed spot, and the route proposal model generation unit generates the route proposal model which outputs a determination of whether or not to make the candidate spot a proposed spot in response to the input of the information of the candidate spot and the user's travel status, and the travel route proposal system according to Configuration 3. According to the travel route suggestion system of configuration 4, a route suggestion model is generated that reflects the user's travel status at the time the information on suggested spots is output, and suggested spots can be determined according to the travel status of the target user using the generated route suggestion model.

[0061] (Configuration 5) The travel route proposal system according to Configuration 3 or Configuration 4, wherein the route proposal model generation unit weights the candidate spots that have been selected as requiring a stopover by the stopover selection reception unit to be extracted as proposed spots with priority over the candidate spots that have been selected as not requiring a stopover by the stopover selection reception unit, and generates the route proposal model by machine learning. According to the travel route suggestion system of Configuration 5, candidate spots that are similar to suggested spots that users have previously selected as necessary stops can be prioritized as suggested spots for the target user in the current trip.

[0062] (Configuration 6) A travel route proposal system according to any one of Configurations 3 to 5, wherein when new proposed result data is added to the behavior history information, the route proposal model generation unit retrains the route proposal model corresponding to the class to which the user who is the target of the added proposed result data belongs, using the added proposed result data. According to the travel route suggestion system of Configuration 6, by updating the route suggestion model to reflect the user's recent behavioral history, it is expected that the likelihood of determining suggested spots that match the target user's needs will increase.

[0063] (Configuration 7) A travel route proposal system according to any one of Configurations 1 to 6, comprising a desired arrival time recognition unit that recognizes a desired arrival time which is the time when the target user will arrive at the destination, and a proposed spot determination unit that extracts spots that the target user can visit and that allow them to reach the destination by the desired arrival time as candidate spots. According to the travel route suggestion system of Configuration 7, candidate spots can be extracted so that the user can arrive at their destination by their desired arrival time, according to their schedule.

[0064] (Configuration 8) A method for proposing a travel route, performed by a computer, comprising: an action history information acquisition step of acquiring action history information indicating the action history of multiple users; a user clustering step of classifying the multiple users into multiple classes based on the action history information; a destination recognition step of recognizing a destination set by a target user who is one of the multiple users; a user location recognition step of recognizing the current location of the target user; a map information acquisition step of acquiring map information of a target area including the destination and the current location of the target user; a proposed spot determination step of extracting candidate spots that can be passed through when traveling from the current location of the target user to the destination based on the map information, and determining whether or not to make the candidate spots proposed spots that the target user should visit, based on the action history information of the users belonging to the target class to which the target user belongs; and a proposed spot information output step of outputting information about the proposed spots. By executing the travel path proposal method of configuration 8 using a computer, the same effects and advantages as the travel path proposal method of configuration 1 can be obtained.

[0065] (Configuration 9) A program that causes a computer to function as: an action history information acquisition unit that acquires action history information showing the action history of multiple users; a user clustering unit that classifies the multiple users into multiple classes based on the action history information; a destination recognition unit that recognizes a destination set by a target user who is one of the multiple users; a user location recognition unit that recognizes the current location of the target user; a map information acquisition unit that acquires map information of a target area including the destination and the current location of the target user; a suggested spot determination unit that extracts candidate spots that the target user can pass through when moving from the current location to the destination based on the map information, and determines whether or not to make the candidate spots suggested spots that the target user should visit based on the action history information of the user belonging to the target class to which the target user belongs; and a suggested spot information output unit that outputs information about the suggested spots. By executing the program of configuration 9 on a computer, the configuration of the travel path proposal system of configuration 1 can be realized. [Explanation of Symbols]

[0066] 1... Travel route suggestion system, 10... Processor, 11... User location recognition unit, 12... Destination recognition unit, 13... Desired arrival time recognition unit, 14... Map information acquisition unit, 15... Behavioral history information acquisition unit, 16... Route suggestion model generation unit, 17... Route suggestion model, 18... Suggested spot determination unit, 19... Suggested spot information output unit, 20... Stopover selection acceptance unit, 21... Suggested route setting unit, 22... User clustering unit, 30... Storage device, 31... Control program for travel route suggestion system, 32... Machine learning platform program, 33...User DB, 34...Behavior history DB, 40...Communication unit, 50...Vehicle, 51...Communication terminal, 60...Communication terminal, 70, 71...Restaurant, 100...Communication network, 110...Traffic information server, 120...Spot information server, 130...Vehicle support server, 200...Suggestion result data, 201...Registered user data, 202...Visit history data, U1, U2...User, Ps...(User's) current location, Pd...Destination.

Claims

1. A behavioral history information acquisition unit acquires behavioral history information showing the behavioral history of multiple users, A user clustering unit that classifies multiple users into multiple classes based on their behavioral history information, A destination recognition unit that recognizes a destination set by a target user who is one of the multiple users, A user location recognition unit that recognizes the current location of the target user, A map information acquisition unit acquires map information of a target area including the destination and the current location of the target user. A suggested spot determination unit extracts candidate spots that the target user can pass through when traveling from the target user's current location to the destination, based on the map information, and determines whether or not to designate the candidate spots as suggested spots to visit for the target user, based on the behavioral history information of the user belonging to the target class to which the target user belongs. A proposed spot information output unit that outputs information about the aforementioned proposed spot, A travel route proposal system equipped with the following features.

2. A stopover selection reception unit receives a selection from the target user regarding whether or not they need to stop at the suggested spot, based on the output of information about the suggested spot by the suggested spot information output unit. The proposed route setting unit, when the stopover selection reception unit receives a selection indicating that a stopover at the proposed spot is necessary, sets a proposed route from the user's current location to the destination via the proposed spot, based on the map information. A travel route proposal system according to claim 1.

3. The aforementioned activity history information includes proposal result data that associates the information of the proposed spot with the results of the selection by the stop selection reception unit regarding whether or not to stop at the proposed spot. The system includes a route proposal model generation unit that generates a route proposal model for each of the multiple classes, which outputs a determination of whether or not to select a candidate spot as a proposed spot in response to input information about the candidate spot, by performing machine learning using training data based on the proposed result data for the users belonging to the class. The proposed spot determination unit determines whether or not to designate a candidate spot as a proposed spot using the route proposal model corresponding to the target class. The travel route proposal system according to claim 2.

4. The aforementioned activity history information includes the proposed spot information, the proposed spot information output unit's proposed spot information output, the proposed spot information output unit's proposed spot information, the proposed spot information output unit's proposed spot information, the proposed spot information output unit's proposed spot information, and the proposed spot information output unit's proposed spot selection result data, which associates these with the proposed spot information output unit's proposed spot information with the proposed spot information output unit's proposed spot selection result. The route proposal model generation unit generates a route proposal model that outputs a determination of whether or not to select a candidate spot as a proposed spot, based on the input of information on the candidate spots and the user's movement status. The travel route proposal system according to claim 3.

5. The route proposal model generation unit assigns weights to candidate spots that have been selected as requiring a stop by the stop selection reception unit, prioritizing them over candidate spots that have been selected as not requiring a stop by the stop selection reception unit, and generates the route proposal model using machine learning. A travel route proposal system according to claim 3 or claim 4.

6. When new proposed result data is added to the behavior history information, the route proposal model generation unit uses the added proposed result data to retrain the route proposal model corresponding to the class to which the user who is the target of the added proposed result data belongs. A travel route proposal system according to claim 3 or claim 4.

7. The system includes a desired arrival time recognition unit that recognizes the desired arrival time, which is the time of arrival at the destination desired by the target user. The proposed spot determination unit extracts as candidate spots spots that the target user can visit and that allow them to reach the destination by the desired arrival time. A travel route proposal system according to any one of claims 1 to 4.

8. A method for proposing travel routes performed by a computer, A step of acquiring behavioral history information for multiple users, which shows the behavioral history of the said users, A user clustering step that classifies multiple users into multiple classes based on their behavioral history information, A destination recognition step that recognizes a destination set by a target user who is one of the multiple users, A user location recognition step that recognizes the current location of the target user, A map information acquisition step involves acquiring map information of a target area that includes the destination and the current location of the target user. A suggested spot determination step involves extracting candidate spots that the target user can pass through when traveling from their current location to the destination, based on the map information, and determining whether or not to designate the candidate spots as suggested spots to visit for the target user, based on the behavioral history information of the user belonging to the target class to which the target user belongs. A suggested spot information output step that outputs information about the aforementioned suggested spot, A method for proposing travel routes, including the provision of a travel route.

9. Computers, A behavioral history information acquisition unit acquires behavioral history information showing the behavioral history of multiple users, A user clustering unit that classifies multiple users into multiple classes based on their behavioral history information, A destination recognition unit that recognizes a destination set by a target user who is one of the multiple users, A user location recognition unit that recognizes the current location of the target user, A map information acquisition unit acquires map information of a target area including the destination and the current location of the target user. A suggested spot determination unit extracts candidate spots that the target user can pass through when traveling from the target user's current location to the destination, based on the map information, and determines whether or not to designate the candidate spots as suggested spots to visit for the target user, based on the behavioral history information of the user belonging to the target class to which the target user belongs. A proposed spot information output unit that outputs information about the aforementioned proposed spot, A program that makes something work.

Images