Movement path suggestion system, movement path suggestion method, and recording medium

Abstract

The invention provides a movement path suggestion system, a movement path suggestion method and a recording medium. When a destination is set by a user, a passing place which is highly likely to coincide with the demand of the user is provided. A movement route recommendation system (1) is provided with: a user clustering unit (22) that classifies a plurality of users into a plurality of categories on the basis of behavior history information of the users; a map information acquisition unit (14) that acquires map information of a target region including the destination set by the target user and the current position of the target user; and a suggested location determination unit (18) that extracts, on the basis of the map information, candidate locations that may be passed when moving from the current location of the target user to the destination, and determines, on the basis of the behavior history information of the users belonging to the target category, whether or not to set the candidate locations as suggested locations on which the target user is suggested to go forward. The object category is a category to which the object user belongs; and a suggested location information output unit (19) that outputs information on the suggested location.

Description

Technical Field

[0001] This invention relates to a mobility route suggestion system, a mobility route suggestion method, and a recording medium for suggesting mobility routes to a destination. Background Technology

[0002] In recent years, there has been an active effort to provide access to sustainable transportation systems to vulnerable groups among transportation participants, including the elderly, people with disabilities, and children. To achieve this, research and development related to navigation technologies is being undertaken to further improve the safety and convenience of transportation.

[0003] For example, Patent Document 1 discloses an action plan provision method that provides action plan information to a user based on the estimated dwell time of the user at a predetermined location estimated according to the user's action information and the user's preference information. Additionally, Patent Document 2 discloses an information processing device that provides information on the next destination to a user based on the user's dwell time at the destination.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2011-60059

[0007] Patent Document 2: Japanese Patent Application Publication No. 2019-124605 Summary of the Invention

[0008] The problem that the invention aims to solve

[0009] However, in navigation technology, in order to enable users to comfortably move to their set destination, it is considered to provide transit points that meet the user's needs. However, the technologies described in Patent Documents 1 and 2 above provide the user with the destination, and therefore cannot provide transit points that meet the user's needs based on the destination set by the user. Therefore, the object of this application is to provide transit points that meet the user's needs when the user has set a destination.

[0010] This application addresses the aforementioned issues by providing a route suggestion system and method that, when a user sets a destination, offers transit points highly likely to align with the user's needs. Furthermore, this contributes to the development of sustainable transportation systems.

[0011] Methods for solving problems

[0012] As a first aspect of the present application, there is provided a mobile route suggestion system including: an action history information acquisition unit configured to acquire, for a plurality of users, action history information indicating action histories of the users; a user clustering unit configured to cluster the plurality of users into a plurality of categories based on the action history information; a destination identification unit configured to identify a destination set by a target user who is any of the plurality of users; a user position identification unit configured to identify a current position of the target user; a map information acquisition unit configured to acquire map information of a target area including the destination and the current position of the target user; a suggestion point decision unit configured to extract, based on the map information, a candidate point that can be passed through when moving from the current position of the target user to the destination, and decide whether or not to set the candidate point as a suggestion point through which the target user is suggested to pass by, based on the action history information of the users belonging to a category to which the target user belongs; and a suggestion point information output unit configured to output information of the suggestion point.

[0013] In the mobile route suggestion system described above, the mobile route suggestion system can further include: a pass-by selection reception unit configured to receive a selection of whether or not to pass by the suggestion point by the target user in response to the output of the information of the suggestion point by the suggestion point information output unit; and a suggestion route setting unit configured to set a suggestion route from the current position of the user to the destination via the suggestion point based on the map information when the selection of passing by the suggestion point is received by the pass-by selection reception unit.

[0014] In the mobile route suggestion system described above, the mobile route suggestion system can further include a suggestion result data in the action history information, the suggestion result data corresponding the information of the suggestion point and a reception result of the selection of whether or not to pass by the suggestion point by the pass-by selection reception unit, and the mobile route suggestion system can further include a route suggestion model generation unit configured to generate a route suggestion model for each of the plurality of categories by machine learning using teacher data based on the suggestion result data related to the users belonging to the category for each of the plurality of categories, the route suggestion model outputting a decision of whether or not to set the candidate point as the suggestion point in response to an input of the information of the candidate point, and the suggestion point decision unit can decide whether or not to set the candidate point as the suggestion point using the route suggestion model corresponding to the target category.

[0015] In the above mobile route suggestion system, the action history information can include the suggestion point information and the suggestion result data corresponding to the suggestion result of whether or not the user desires to go to the suggestion point, the suggestion result being obtained by the on-the-way selection receiving unit receiving the selection of whether or not the user desires to go to the suggestion point, and the route suggestion model generating unit can generate the route suggestion model that outputs the determination of whether or not to set the candidate point as the suggestion point in response to the input of the candidate point information and the user's movement status.

[0016] In the above mobile route suggestion system, the route suggestion model generating unit can generate the route suggestion model by the machine learning, in which the candidate points for which the on-the-way selection receiving unit receives the selection of going on the way are weighted so as to be extracted as the suggestion points in preference to the candidate points for which the on-the-way selection receiving unit receives the selection of not going on the way.

[0017] In the above mobile route suggestion system, the route suggestion model generating unit can perform the relearning of the route suggestion model corresponding to the category to which the user belongs, who is the object of the suggestion result data, using the newly added suggestion result data when the new suggestion result data is added to the action history information.

[0018] In the above mobile route suggestion system, the mobile route suggestion system can include a desired arrival time identifying unit that identifies a desired arrival time at which the object user desires to arrive at the destination, and the suggestion point determining unit can extract, as the candidate point, a point at which the object user can arrive at the destination at or before the desired arrival time after going on the way to the point.

[0019] As a second aspect for achieving the above object, there is provided a movement route suggestion method executed by a computer, the movement route suggestion method including: an action history information acquisition step of acquiring, for a plurality of users, action history information indicating action histories of the users; a user clustering step of clustering the plurality of users into a plurality of categories based on the action history information; a destination identification step of identifying a destination set by an object user who is any of the plurality of users; a user position identification step of identifying a current position of the object user; a map information acquisition step of acquiring map information of an object area including the destination and the current position of the object user; a suggestion point decision step of extracting, based on the map information, a candidate point that is likely to be passed through when moving from the current position of the object user to the destination, and deciding whether to set the candidate point as a suggestion point for the object user to pass by on the way to the destination based on the action history information of the users belonging to a category to which the object user belongs, which is an object category; and a suggestion point information output step of outputting information of the suggestion point.

[0020] As a third aspect for achieving the above object, there is provided a recording medium recording a program causing a computer to function as: an action history information acquisition section acquiring, for a plurality of users, action history information indicating action histories of the users; a user clustering section clustering the plurality of users into a plurality of categories based on the action history information; a destination identification section identifying a destination set by an object user who is any of the plurality of users; a user position identification section identifying a current position of the object user; a map information acquisition section acquiring map information of an object area including the destination and the current position of the object user; a suggestion point decision section extracting, based on the map information, a candidate point that is likely to be passed through when moving from the current position of the object user to the destination, and deciding whether to set the candidate point as a suggestion point for the object user to pass by on the way to the destination based on the action history information of the users belonging to a category to which the object user belongs, which is an object category; and a suggestion point information output section outputting information of the suggestion point.

[0021] Effects of Invention

[0022] According to the above movement route suggestion system, movement route suggestion method, and recording medium, when a user sets a destination, it is possible to provide a passing point with high possibility that is consistent with the needs of the user. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a configuration diagram of the movement route suggestion system in the present embodiment.

[0024] Figure 2 is an explanatory diagram of data held in the storage device.

[0025] Figure 3 is an explanatory diagram of a cluster of a plurality of users.

[0026] Figure 4 is an explanatory diagram of a generation process of a route suggestion model.

[0027] Figure 5 is a first flowchart of a process of suggesting a detour to a suggestion point.

[0028] Figure 6 is a second flowchart of a process of suggesting a detour to a suggestion point.

[0029] Figure 7 is an explanatory diagram of a series of processes from the determination of a suggestion point to the saving of a selection result of a detour to the suggestion point.

[0030] BRIEF DESCRIPTION OF DRAWINGS

[0031] 1: mobile route suggestion system, 10: processor, 11: user position recognition unit, 12: destination recognition unit, 13: desired arrival time recognition unit, 14: map information acquisition unit, 15: action history information acquisition unit, 16: route suggestion model generation unit, 17: route suggestion model, 18: suggestion point determination unit, 19: suggestion point information output unit, 20: detour selection reception unit, 21: suggestion route setting unit, 22: user clustering unit, 30: storage device, 31: control program of the mobile route suggestion system, 32: platform program of machine learning, 33: user DB, 34: action history DB, 40: communication unit, 50: vehicle, 51: communication terminal, 60: communication terminal, 70, 71: restaurant, 100: communication network, 110: traffic information server, 120: point information server, 130: vehicle support server, 200: suggestion result data, 201: registered user data, 202: visit result data, U1, U2: user, Ps: current position (of the user), Pd: destination. DETAILED DESCRIPTION

[0032] [1. Structure of the mobile route suggestion system]

[0033] REFERENCE Figure 1The structure of the mobile route suggestion system 1 of the present embodiment will be described. The mobile route suggestion system 1 is a computer system connected to the communication network 100. The mobile route suggestion system 1 provides information on places that can be visited along the way to the destination in the middle of the movement to the user by communicating with the communication terminal used by the user of the mobile route suggestion system 1 via the communication network 100.

[0034] In Figure 1 , as the user of the mobile route suggestion system 1, the user U1 who moves by riding the vehicle 50 and the user U2 who moves on foot are exemplified. In Figure 1 , Ps is the current position of the users U1, U2, and Pd is the destination of the users U1, U2. The mobile route suggestion system 1 provides information on places that can be visited along the way to the user U1 by communicating with the communication terminal 51 (navigation device, DA (Display Audio), smartphone, tablet terminal, portable telephone, etc.) equipped in the vehicle 50 or brought into the vehicle 50 by the user U1. In addition, the mobile route suggestion system 1 provides information on places that can be visited along the way to the user U2 by communicating with the communication terminal 60 (smartphone, tablet terminal, portable telephone, etc.) carried by the user U2.

[0035] Further, the mobile route suggestion system 1 communicates with various servers such as the traffic information server 110, the place information server 120, and the vehicle support server 130 via the communication network 100. The traffic information server 110 provides traffic information such as a map, road regulation, etc. The place information server 120 provides information on places that can be visited along the way (restaurants, service areas, drive-through service facilities, sightseeing places, etc.). The vehicle support server 130 receives and stores information on the movement status (movement place, movement time, movement speed, etc.) from the service object vehicle such as the vehicle 50.

[0036] The mobile route suggestion system 1 has a processor 10 (corresponding to the computer of the present disclosure), a storage device 30, a communication unit 40, etc. In the storage device 30, a control program 31 (corresponding to the program of the present disclosure) of the mobile route suggestion system 1, a platform program 32 of machine learning, a user DB (database) 33, and a behavior history DB 34 are stored. As Figure 2 shown, in the user DB 33, the user IDs (User01, User02,...) individually issued to a plurality of users registered in the mobile route suggestion system 1, the categories to which the users belong (in the present embodiment, the user U1 belongs to the category of the user who moves by riding the vehicle 50, and the user U2 belongs to the category of the user who moves on foot), and the like are recorded. Figure 2The registration user data 201 is associated with the category of the user (shown as a user category) and the registration information (communication address, home address, preferences, etc.) input by each user at the time of registration. The category to which the user belongs is determined by the clustering process described later.

[0037] In the action history DB 34, the Figure 2 suggested result data 200 and the visit result data 202 shown in FIG. 6 are recorded for each user. The suggested result data 200 and the visit result data 202 correspond to the action history information of the present disclosure. In the suggested result data 200, for each suggestion of a stop-by suggestion location made to the user, the date and time at which the suggestion was made, the selection result of the user as to whether or not the stop-by suggestion location was needed (needed or not needed), the information of the suggestion location (suggestion division, category), and the movement situation of the user at the time of making the suggestion (driving time, movement distance, distance from home, number of stops on the day) (suggested-time movement situation) are included. In the visit result data 202, the locations visited by the user up to the present time are divided by genre, and the number of visits is recorded. The visit result data 202 is used for the clustering of users, the estimation of the preferences of the user with respect to locations, and the like described later.

[0038] The processor 10 functions as the user position recognition section 11, the destination recognition section 12, the desired arrival time recognition section 13, the map information acquisition section 14, the action history information acquisition section 15, the user clustering section 22, the route suggestion model generation section 16, the suggestion location decision section 18, the suggestion location information output section 19, the stop-by selection reception section 20, and the suggestion route setting section 21 by reading in and executing the control program 31 of the movement route suggestion system 1.

[0039] Here, the processing performed by the user position recognition section 11 corresponds to the user position recognition step in the movement route suggestion method of the present disclosure, and the processing performed by the destination recognition section 12 corresponds to the destination recognition step in the movement route suggestion method of the present disclosure. The processing performed by the map information acquisition section 14 corresponds to the map information acquisition step in the movement route suggestion method of the present disclosure, and the processing performed by the action history information acquisition section 15 corresponds to the action history information acquisition step in the movement route suggestion method of the present disclosure. The processing performed by the suggestion location decision section 18 corresponds to the suggestion location decision step in the movement route suggestion method of the present disclosure, and the processing performed by the suggestion location information output section 19 corresponds to the suggestion location information output step in the movement route suggestion method of the present disclosure. The processing performed by the user clustering section 22 corresponds to the user clustering step in the movement route suggestion method of the present disclosure.

[0040] The following describes the scenario where user U1, who is driving vehicle 50 from the current location Ps to the destination Pd, is advised to visit a location along the way. The same applies to other users, such as user U2, who are advised to visit locations along the same route.

[0041] The user location identification unit 11 communicates with the communication terminal 51 to obtain the detection information of the current location of the user U1 (the current location of the vehicle 50) detected by the GNSS (Global Navigation Satellite System) sensor and other location sensors of the communication terminal 51, and identifies the current location of the user U1.

[0042] The destination identification unit 12 identifies the destination Pd set by the user U1 operating the communication terminal 51 by communicating with the communication terminal 51. The desired arrival time identification unit 13 identifies the desired arrival time for destination Pd set by the user U1 operating the communication terminal 51 by communicating with the communication terminal 51.

[0043] The map information acquisition unit 14 obtains map information of the object area containing the current location Ps and destination Pd of the user U1 from the traffic information server 110. The action history information acquisition unit 15 obtains the movement history information of the vehicle 50 by communicating with the communication terminal 51 or accessing the vehicle support server 130. In addition, it identifies the acceptance result of the option selection acceptance unit 20 (described later) on whether to accept the option to go to the suggested location, generates suggestion result data 200 and records it in the action history DB34.

[0044] like Figure 3 As shown, the user clustering unit 22 performs hierarchical clustering for multiple users, using the total number of visits to each category (Italian, Japanese, Chinese, etc.) recorded in the visit outcome data 202 for each user. Then, the user clustering unit 22 calculates the silhouette coefficient and automatically sets the number of clusters that maximize the silhouette coefficient. The category to which each user belongs, as classified by the user clustering unit 22 (in... Figure 3 The data (including examples of shopping purposes and weekend leisure activities) is recorded in the Registered User Data 201.

[0045] like Figure 4 As shown, the path suggestion model generation unit 16 refers to the registered user data 201 (refer to...). Figure 2The suggestion result data 200 stored for the plurality of users is classified by different categories based on the category of each user recorded in the user information table 300. Then, the route suggestion model generation section 16 generates a route suggestion model 17 corresponding to each category using the suggestion result data 200 classified by different categories.

[0046] Figure 4 A case where a route suggestion model corresponding to the category of "shopping use" is generated is exemplified. The route suggestion model generation section 16 generates teacher data 210 in which the information of the suggestion place and the movement situation are exemplified as the test questions and the selection result is the correct answer, by performing dummy variable processing on the suggestion result data 200 of the users belonging to the category of "shopping use", in which the subdivision of the suggestion date and time (weekday, holiday, day of the week, time zone) and the subdivision of the attributes of the suggestion place (style division, purpose division, etc.) are performed in the dummy variable processing.

[0047] Then, the route suggestion model generation section 16 executes the platform program 32 of machine learning, and performs machine learning using the teacher data 210, whereby a route suggestion model 17 corresponding to the category of "shopping use" is generated as shown in the dialog box B1, which outputs the determination of whether or not the candidate place is suggested as the suggestion place with respect to the input of the information of the candidate place and the movement situation of the user. The route suggestion model generation section 16 performs the same processing for other categories, whereby the route suggestion models 17 corresponding to each category are generated respectively.

[0048] Here, the route suggestion model generation section 16 can perform weighting so that the candidate place for which the selection by the on-the-way selection accepting section 20 is the on-the-way selection is preferentially extracted as the suggestion place compared to the candidate place for which the selection by the on-the-way selection accepting section 20 is not the on-the-way selection, when generating the route suggestion model 17 by machine learning.

[0049] The suggestion place decision section 18 accesses the place information server 120, and acquires the information of the place existing in the vicinity of the user U1, whereby the candidate place to which the user U1 can go on the way to the destination Pd is extracted. In this case, the suggestion place decision section 18 extracts the candidate step with the condition that the destination Pd can be reached at the desired arrival time recognized by the desired arrival time recognition section 13 or before the desired arrival time.

[0050] Then, the suggested location decision unit 18 inputs the candidate location information and the user U1's movement status into the route suggestion model 17, and decides whether to set the candidate location as a suggested location for the user U1 to travel along the same route based on the decision output of the route suggestion model 17. The suggested location information output unit 19 sends the suggested location information to the communication terminal 51, suggesting that the user travel along the same route to the suggested location. The route travel selection acceptance unit 20 receives the selection result data sent from the communication terminal 51 regarding whether to travel along the same route to the suggested location, and accepts the selection of the suggested location to travel along the same route.

[0051] When the route suggestion setting unit 21 receives a request from the route selection receiving unit 20 to travel to a suggested location, it sets a suggested route from the user U1's current location (the current location of the vehicle 50) to the destination Pd via the suggested location, and sends 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 the screen and executes route guidance according to the user U1's instructions.

[0052] [2. Suggestions for visiting the suggested locations along the way]

[0053] according to Figure 5 , Figure 6 The flowchart shown is for reference. Figure 7 The outline of a series of processes shown is used to illustrate the process in Figure 1 The steps shown illustrate the process of suggesting a route to a suggested location performed by the mobility path suggestion system 1 for user U1 under the indicated conditions. In this case, user U1 is equivalent to the user object of this disclosure.

[0054] When the movement path suggestion system 1 identifies, through communication with the communication terminal 51, that user U1 has started moving using vehicle 50, it executes the following steps: Figure 5 , Figure 6 The flowchart shows the suggested processing for going to the suggested location along the way. Figure 5 In step S1, the mobile path suggestion system 1 identifies the user ID of user U1 by communicating with the communication terminal 51, and determines user U1.

[0055] In the next step S2, the destination identification unit 12 determines whether a destination has been set for user U1 through communication with the communication terminal 51. If a destination has been set, the process proceeds to step S3; otherwise, the process proceeds to step S20. In step S20, the destination identification unit 12 determines whether user U1 has finished moving using vehicle 50 through communication with the communication terminal 51. Then, the destination identification unit 12 initiates the process when user U1 has finished moving. Figure 6to step S2.

[0056] In step S3, the user position recognition section 11 recognizes the current position of the user U1 (the current position of the vehicle 50) by communication with the communication terminal 51. In the next step S4, the map information acquisition section 14 acquires map information and road traffic information of the subject region including the current position Ps and the destination Pd of the user U1 from the traffic information server 110. In the next step S5, the desired arrival time recognition section 13 recognizes the desired arrival time at the destination Pd set by the user U1 by communication with the communication terminal 51.

[0057] In the next step S6, the suggestion site decision section 18 refers to the map information and the traffic information to search for a site that the user U1 can visit on the way from the current position Ps to the destination Pd, i.e., a candidate site, on the condition that the destination Pd can be reached by the desired arrival time or before the desired arrival time. Here, the suggestion site decision section 18 assumes a usual stay time at the site according to the style of the site (a restaurant, a service area, a drive-through service facility, a convenience store, etc.) and searches for a candidate site in which the destination Pd can be reached by the desired arrival time or before the desired arrival time in the case of visiting on the way.

[0058] In the next step S7, the suggestion site decision section 18 makes the process proceed to S7 when the candidate site is extracted and makes the process proceed to Figure 6 In step S18, the suggestion site decision section 18 ends the present on-the-way visit suggestion process. In step S8, as shown in the dialog box B2, the suggestion site decision section 18 recognizes the movement situation of the user U1 (the movement situation of the vehicle 50) based on the movement history data recognized by the movement history information acquisition section 15 and saved in the movement history DB 34. Figure 7

[0059] In the next step S9, the suggestion site decision section 18 creates inference data 220 shown in FIG. 9 based on the information of the candidate site and the movement situation of the user U1. The inference data 220 includes the category to which the user U1 belongs (here, "shopping purpose"), the information of the candidate site (suggestion division, classification), and the movement situation of the user U1 (time period, driving time, movement distance, number of stops on the day). In the next step S10, the suggestion site decision section 18 inputs the inference data 220 to the route suggestion model 17 corresponding to the category of "shopping purpose" as shown in the dialog box B3 and acquires the output of the inference result (suggestion / non-suggestion). Figure 7 Figure 7

[0060] ​​​Further, in a case where the route suggestion model 17 targeted at the user U1 has not been generated due to insufficient collection of the suggestion result data 200 (refer to Figure 2 ) related to the user U1, the suggestion place decision section 18 decides a candidate place of a style that matches the preference of the user U1 recorded in the registered user data 201 (refer to Figure 2 ) as the suggestion place, for example.

[0061] In the next step Sll, the suggestion place decision section 18 causes the process to proceed to step S12 when the output of the route suggestion model 17 is "suggest", and causes the process to proceed to step S18 of Figure 6 when the output of the route suggestion model 17 is "not suggest". In step S12, the suggestion place information output section transmits suggestion place information suggesting a detour to the suggestion place to the communication terminal 51 as shown in the dialog box B4. Figure 7 As shown in Figure 7 , the communication terminal 51 that received the suggestion place information displays a detour suggestion screen 55 on the display.

[0062] In the detour suggestion screen 55, it is displayed that the restaurant as the suggestion place is located nearby, a "yes" button 55a for selecting that a detour is needed, and a "no" button 55b for selecting that a detour is not needed. The communication terminal 51 transmits selection result information indicating the operation of the user U1 to the "yes" button 55a or the "no" button, or the selection result of "yes" or "no" by the input through the voice to the mobile route suggestion system 1. In the next step S13, the detour selection acceptance section 20 causes the process to proceed to step S14 when receiving the selection result information transmitted from the communication terminal 51. Figure 6

[0063] In step S14, the detour selection acceptance section 20 identifies the selection result of the user U1 as to whether a detour is needed or not to the suggestion place, based on the selection result information. Then, the detour selection acceptance section 20 causes the process to proceed to step S15 when the selection result is that a detour is needed, and causes the process to proceed to step S16 when the selection result is that a detour is not needed. In Figure 7 the example, as shown in the dialog box B5, the user U1 selects that a detour is not needed, and thus goes to the destination Pd (park) without detouring to the Italian restaurant 70.

[0064] In step S15, the suggestion route setting section 21 transmits information of a suggestion route to the destination Pd via the suggestion place to the communication terminal 51. In Figure 1 ​In the example, when restaurant 70 is the suggested location, the suggested route setting unit 21 sends information about the suggested route R1 from restaurant 70 to destination Pd to the communication terminal 51. Additionally, when restaurant 71 is the suggested location, the suggested route setting unit 21 sends information about the suggested route R2 from restaurant 71 to destination Pd to the communication terminal 51.

[0065] Furthermore, in step S12, if no selection result information is received from the communication terminal 51 in step S13 within a specified time from the moment the suggestion information for going to the suggested location is sent to the communication terminal 51, the route selection acceptance unit 20 can determine that the user U1 has chosen not to go to the suggested location and proceed to step S16, or it can proceed to step S18.

[0066] In step S16, the action history information acquisition unit 15, as follows: Figure 7 As shown in dialog box B6, the system generates suggestion result data 200a for this operation, including the selection results of whether or not to go to the suggested location along the way. As shown in dialog box B7, this suggestion result data 200a is appended to the suggestion result data 200 in the action history DB34. In the next step S17, the route suggestion model generation unit 16 determines whether a predetermined number or more of suggestion result data 200 have been appended to the action history DB34, based on the category of "shopping purpose" for the current target process. Then, if a predetermined number or more of suggestion result data 200 have been appended, the route suggestion model generation unit 16 proceeds to step S30; otherwise, it proceeds to step S18.

[0067] In step S30, the path suggestion model generation unit 16 generates a path suggestion model 17 corresponding to the category of "shopping purpose" by referring to the path suggestion model 17, which includes the additional suggestion result data 200. Figure 4 The aforementioned machine learning (relearning) is used to update the path suggestion model 17 corresponding to the "shopping purpose" category. By updating the path suggestion model 17 to reflect the recent action history of users belonging to the "shopping purpose" category, it is expected to increase the likelihood of identifying suggested locations consistent with the needs of users belonging to the "shopping purpose" category.

[0068] [3. Other Implementation Methods]

[0069] In the above-described embodiment, the computer system (server) that communicates with the communication terminal 51 used in the vehicle 50 and the communication terminal 60 carried by the user U2 via the communication network 100, that is, the mobile route suggestion system 1 performs the function of the mobile route suggestion system of the present disclosure. As another embodiment, the communication terminals 51, 60 can perform the function of the mobile route suggestion system of the present disclosure. Alternatively, a part of the mobile suggestion system of the present disclosure can perform the function of the server, and the remaining part can be configured by the communication terminals 51, 60. In this case, the mobile route suggestion system of the present disclosure is configured by the server, the communication terminal 51, or the communication terminal 60. For example, the following configuration can be adopted: the route suggestion model generation section 16 and the suggestion point decision section 18, in which the load of the operation processing is high, are configured in the server, and the other configurations are configured in the communication terminals 51, 60.

[0070] In the above-described embodiment, as shown in Figure 4 , the route suggestion model generation section 16 generates the route suggestion model 17 using the teacher data 210 including the information of the suggestion point, the movement status of the user U1, and the selection result of the user U1 as to whether or not the suggestion point needs to be visited on the way. As another embodiment, teacher data including the information of the suggestion point and the selection result of the user U1 as to whether or not the suggestion point needs to be visited on the way can be used without including the movement status of the user U1.

[0071] In the above-described embodiment, the mobile route suggestion system 1 has the desired arrival time recognition section 13, and the suggestion point decision section 18 searches for the candidate point with the condition that the destination Pd is reached at the desired arrival time or before the desired arrival time. As another embodiment, the mobile route suggestion system 1 can be configured to omit the desired arrival time recognition section 13, and the suggestion point decision section 18 searches for the candidate point without considering the desired arrival time.

[0072] In the above-described embodiment, the route suggestion model generation section 16 performs the relearning of the route suggestion model 17 using the additional suggestion result data as shown in steps S17, S30 of Figure 6 , but can be configured not to perform the relearning of the route suggestion model 17.

[0073] In the above-described embodiment, the route suggestion model generation section 16 is provided, and the suggestion point decision section 18 decides the suggestion point using the route suggestion model 17 generated by the route suggestion model generation section 16. As another configuration, the suggestion point can be decided by another method without using the route suggestion model 17 generated by machine learning, for example, a point that matches the preference or the movement pattern of the user estimated from the action history of the user is decided as the suggestion point, and the like.

[0074] Further, Figure 1This is a schematic diagram illustrating the functional structure of the movement path suggestion system 1 based on its main processing content for ease of understanding of the invention. The functional structure of the movement path suggestion system 1 can also be constructed using other division methods. Furthermore, the processing of each structural element can be performed by one hardware unit or by multiple hardware units. Figure 5 , Figure 6 The processing of each structural element shown can be performed by one program or by multiple programs.

[0075] [4. Structure supported by the above embodiments]

[0076] The above implementation is a specific example of the following structure.

[0077] (Structure 1) A mobile route suggestion system, comprising: an action history information acquisition unit that acquires action history information representing the action history of multiple users; a user clustering unit that classifies the multiple users into multiple categories based on the action history information; a destination identification unit that identifies the destination set by an object user who is any one of the multiple users; a user location identification unit that identifies the current location of the object user; a map information acquisition unit that acquires map information of an object area including the destination and the current location of the object user; a suggested location determination unit that extracts candidate locations that may be passed when moving from the current location of the object user to the destination based on the map information, and decides whether to set the candidate locations as suggested locations for the object user to travel along the same route based on the action history information of the users belonging to the category to which the object user belongs; and a suggested location information output unit that outputs the information of the suggested location.

[0078] According to the movement path suggestion system of Structure 1, when the user sets a destination, it can provide transit points that are highly likely to be consistent with the user's needs based on the user's action history of the user's object category.

[0079] (Structure 2) The mobile path suggestion system according to Structure 1 includes: a route selection acceptance unit that accepts the user's selection of whether to travel to the suggested location based on the output of suggested location information by the suggested location information output unit; and a suggested route setting unit that, when the route selection acceptance unit accepts the selection of whether to travel to the suggested location, sets a suggested route from the user's current location to the destination via the suggested location based on the map information.

[0080] According to the movement route suggestion system of Structure 2, it is possible to set a suggested route to a destination via a suggested place in accordance with a selection of necessity or not by a user.

[0081] (Structure 3) The movement route suggestion system of Structure 2, wherein the action history information includes suggestion result data that corresponds information of the suggested place and a reception result of the reception of the selection of the necessity or not of the on-the-way visit to the suggested place by the on-the-way visit selection reception unit, and the route suggestion model generation unit generates a route suggestion model for each of the plurality of categories by machine learning using teacher data based on the suggestion result data related to the user belonging to the category, the route suggestion model outputs a determination of whether or not to set the candidate place as the suggested place for an input of information of the candidate place, and the suggested place decision unit decides whether or not to set the candidate place as the suggested place using the route suggestion model corresponding to the object category.

[0082] According to the movement route suggestion system of Structure 3, it is possible to reflect a selection result of the necessity or not of the on-the-way visit to the suggested place by the user belonging to each category in the past for each of the plurality of categories, generate a route suggestion model corresponding to each category, and decide a suggested place for the object user using the route suggestion model corresponding to the object category to which the object user belongs.

[0083] (Structure 4) The movement route suggestion system of Structure 3, wherein the action history information includes the suggestion result data that corresponds the information of the suggested place and a movement situation of the user at a time when the information of the suggested place is output by the suggested place information output unit, that is, a suggestion-time movement situation, and a reception result of the reception of the selection of the necessity or not of the on-the-way visit to the suggested place by the on-the-way visit selection reception unit, and the route suggestion model generation unit generates the route suggestion model that outputs a determination of whether or not to set the candidate place as the suggested place for an input of the information of the candidate place and the movement situation of the user.

[0084] According to the movement route suggestion system of Structure 4, it is possible to generate a route suggestion model that reflects a movement situation of the user at a time when the information of the suggested place is output, and decide a suggested place corresponding to a movement situation of the object user using the generated route suggestion model.

[0085] (structure 5) The movement route suggestion system according to structure 3 or structure 4, wherein the route suggestion model generation section generates the route suggestion model by the machine learning with weighting that gives priority to the candidate place for which the selection received by the en route selection reception section is the candidate place for which the en route trip is needed, as compared to the candidate place for which the selection received by the en route selection reception section is the candidate place for which the en route trip is not needed.

[0086] The movement route suggestion system according to structure 5 can prioritize the candidate place that is similar to the suggestion place for which the selection by the user in the past is the en route trip, as the suggestion place for the subject user this time.

[0087] (structure 6) The movement route suggestion system according to any one of structures 3 to 5, wherein, when a new suggestion result data is added to the action history information, the route suggestion model generation section performs relearning of the route suggestion model corresponding to the category to which the user who is the subject of the added suggestion result data belongs, using the added suggestion result data.

[0088] The movement route suggestion system according to structure 6 can improve the possibility of determining the suggestion place that matches the needs of the subject user by updating the route suggestion model reflecting the recent action history of the user.

[0089] (structure 7) The movement route suggestion system according to any one of structures 1 to 6, wherein the movement route suggestion system has a desired arrival time recognition section that recognizes a desired arrival time at which the subject user desires to arrive at the destination, and the suggestion place decision section extracts, as the candidate place, a place at which the subject user can arrive at the destination at or before the desired arrival time after making an en route trip to the place.

[0090] The movement route suggestion system according to structure 7 can extract the candidate place in such a manner that the subject user can arrive at the destination at or before the desired arrival time, according to the schedule of the subject user.

[0091] (structure 8) A mobile route suggestion method executed by a computer, including: an action history information obtaining step of obtaining, for a plurality of users, action history information indicating action histories of the users; a user clustering step of clustering the plurality of users into a plurality of categories based on the action history information; a destination identifying step of identifying a destination set by an object user who is any of the plurality of users; a user position identifying step of identifying a current position of the object user; a map information obtaining step of obtaining map information of an object area including the destination and the current position of the object user; a suggestion place deciding step of extracting, based on the map information, a candidate place that is likely to be passed through when moving from the current position of the object user to the destination, and deciding whether to set the candidate place as a suggestion place where the object user is suggested to pass by on the way based on the action history information of the users belonging to a category to which the object user belongs, i.e., an object category; and a suggestion place information output step of outputting information of the suggestion place.

[0092] By executing the mobile route suggestion method of structure 8 by a computer, the same operational effects as the mobile route suggestion method of structure 1 can be obtained.

[0093] (structure 9) A recording medium recording a program causing a computer to function as: an action history information obtaining section obtaining, for a plurality of users, action history information indicating action histories of the users; a user clustering section clustering the plurality of users into a plurality of categories based on the action history information; a destination identifying section identifying a destination set by an object user who is any of the plurality of users; a user position identifying section identifying a current position of the object user; a map information obtaining section obtaining map information of an object area including the destination and the current position of the object user; a suggestion place deciding section extracting, based on the map information, a candidate place that is likely to be passed through when moving from the current position of the object user to the destination, and deciding whether to set the candidate place as a suggestion place where the object user is suggested to pass by on the way based on the action history information of the users belonging to a category to which the object user belongs, i.e., an object category; and a suggestion place information output section outputting information of the suggestion place.

[0094] By executing the program recorded by the recording medium of structure 9 by a computer, the structure of the mobile route suggestion system of structure 1 can be realized.

Claims

1. A movement route suggestion system, which has the following features: The Action History Information Acquisition Unit acquires action history information representing the action history of multiple users. The user clustering department classifies multiple users into multiple categories based on the action history information; A destination identification unit identifies the destination set by an object user, which is any one of the plurality of users. A user location identification unit identifies the current location of the user of the object. The map information acquisition unit acquires map information of the object area, including the destination and the current location of the object user; The suggested location determination unit extracts candidate locations that the user may pass through when moving from the current location of the target user to the destination based on the map information, and decides whether to set the candidate locations as suggested locations for the target user to go along the way based on the user's action history information belonging to the category to which the target user belongs. as well as The suggested location information output unit outputs information about the suggested location.

2. The mobility path suggestion system according to claim 1, comprising: The system allows users to select whether or not they need to visit a suggested location, based on the information output by the suggested location information output unit. The suggested route setting unit, when the route selection processing unit accepts the selection of a route to the suggested location, sets a suggested route from the user's current location to the destination via the suggested location based on the map information.

3. The movement path suggestion system according to claim 2, wherein, The action history information includes suggestion result data, which is obtained by correlating the information of the suggested location with the acceptance results of the "going along the way to the service center" selection whether it is necessary to go to the suggested location. The mobile route suggestion system includes a route suggestion model generation unit. This unit generates a route suggestion model for each of the multiple categories by performing machine learning using teacher data for each of the categories. The teacher data is based on suggestion result data related to the user belonging to the category. The route suggestion model takes as input information about the candidate locations and outputs a determination of whether to set the candidate location as the suggested location. The suggested location decision unit uses the path suggestion model corresponding to the object category to decide whether to set the candidate location as the suggested location.

4. The movement path suggestion system according to claim 3, wherein, The action history information includes the user's movement status at the time the suggested location information and the suggested location information output unit output the suggested location information, i.e., the movement status at the time of suggestion, and the suggestion result data obtained by correlating the suggested location information with the acceptance result of the "going along the way" selection acceptance unit in choosing whether to go to the suggested location. The route suggestion model generation unit generates the route suggestion model, which, based on the information of the candidate locations and the user's movement status, outputs a determination on whether to set the candidate locations as suggested locations.

5. The movement path suggestion system according to claim 3 or 4, wherein, The route suggestion model generation unit generates the route suggestion model by performing the following weighting on the candidate locations selected by the route selection service unit as places that need to be visited along the way, and then using the machine learning to generate the route suggestion model. This weighting makes the candidate locations selected by the route selection service unit as places that need to be visited along the way preferred over the candidate locations selected by the route selection service unit as places that do not need to be visited along the way.

6. The movement path suggestion system according to claim 3 or 4, wherein, When new suggestion result data is added to the action history information, the path suggestion model generation unit uses the added suggestion result data to relearn the path suggestion model corresponding to the category to which the user belongs, and the user is the object of the added suggestion result data.

7. The mobility path suggestion system according to any one of claims 1 to 4, wherein, The mobility path suggestion system has a desired arrival time identification unit, which identifies the desired arrival time of the user to reach the destination. The suggested location determination unit extracts the following locations as candidate locations, so that the user can reach the destination at or before the desired arrival time after going to the location along the way.

8. A method for suggesting a movement path, executed by a computer, comprising: The action history information acquisition step involves acquiring action history information representing the action history of multiple users. The user clustering step classifies multiple users into multiple categories based on the action history information; The destination identification step identifies the destination set by an object user who is any one of the multiple users. The user location identification step identifies the current location of the user of the object. The map information acquisition step involves acquiring map information of the object area, including the destination and the current location of the object user. The suggested location determination step involves extracting candidate locations that the user might pass through when moving from the current location to the destination based on the map information, and determining whether to set the candidate locations as suggested locations for the user to go along the way based on the user's action history information belonging to the category to which the user belongs, i.e., the object category. as well as The suggested location information output step outputs the information of the suggested location.

9. A recording medium having recorded a program that enables a computer to function as a component of: The Action History Information Acquisition Unit acquires action history information representing the action history of multiple users. The user clustering department classifies multiple users into multiple categories based on the action history information; A destination identification unit identifies the destination set by an object user, which is any one of the plurality of users. A user location identification unit identifies the current location of the user of the object. The map information acquisition unit acquires map information of the object area, including the destination and the current location of the object user; The suggested location determination unit extracts candidate locations that the user may pass through when moving from the current location of the target user to the destination based on the map information, and decides whether to set the candidate locations as suggested locations for the target user to go along the way based on the user's action history information belonging to the category to which the target user belongs. as well as The suggested location information output unit outputs information about the suggested location.

Images