Navigation recommendation method and device

Abstract

The invention provides a navigation recommendation method and device, and the method comprises the steps: responding to a navigation request carrying destination information of a target object, and obtaining the historical travel data of the target object, the current position of the target object and the object attribute information; according to the historical travel data and the object attribute information, analyzing the destination information, and determining the travel intention of the target object; based on the travel intention, performing destination recommendation processing on the destination information, generating candidate recommendation destinations, and sending the candidate recommendation destinations to the navigation terminal, so that the navigation terminal displays the candidate recommendation destinations; acquiring a current navigation destination sent by the navigation terminal; and based on the object attribute information and the object navigation preference model, carrying out route recommendation processing on the current navigation destination and the current position, generating candidate recommendation routes, and sending the candidate recommendation routes to the navigation terminal, so that all-around and personalized travel service recommendation can be provided for the user.

Description

Technical Field

[0001] The present application relates to the field of navigation technology, and in particular to a navigation recommendation method and device. Background Art

[0002] Navigation software is widely used in daily life and work, providing users with services such as route planning and location search. However, in the prior art, when navigation software recommends location search results and plans routes, most navigation systems only plan routes based on the starting point, end point and real-time traffic conditions, and lack consideration of users' personalized travel preferences, such as preference for scenic roads; and the recommendation service is single: only destination recommendations are provided, and there are few recommendations for related services along the way or in the surrounding area, such as catering, parking, attractions, etc.; and the situational awareness ability is limited: it fails to fully utilize users' real-time situational information to optimize the recommended content, such as time, weather, vehicle type, etc. Therefore, there is an urgent need for a navigation recommendation method that can make personalized recommendations for destinations and routes based on users' navigation habits and interests, as well as recommend locations along the way. Summary of the invention

[0003] In order to solve the above technical problems, the present application proposes a navigation recommendation method and device.

[0004] On the one hand, an embodiment of the present application provides a navigation recommendation method, the method comprising:

[0005] In response to a navigation request of a target object, acquiring historical travel data of the target object, a current location of the target object, and object attribute information; the navigation request carries destination information;

[0006] Analyzing the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object;

[0007] Based on the travel intention, performing destination recommendation processing on the destination information to generate candidate recommended destinations, and sending the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations;

[0008] Acquire a current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal;

[0009] Based on the object attribute information and the object navigation preference model, route recommendation processing is performed on the current navigation destination and the current location, candidate recommended routes are generated, and the candidate recommended routes are sent to the navigation terminal so that the navigation terminal displays the candidate recommended routes; the object navigation preference model is obtained by processing the historical travel data.

[0010] Furthermore, the method for constructing the object attribute information includes:

[0011] Acquire historical third-party software data of the target object; the historical third-party software data represents historical data of third-party software associated with the navigation terminal;

[0012] Cleaning the historical travel data to remove invalid data in the historical travel data to obtain historical travel information;

[0013] Classifying and filtering the historical third-party software data to determine the historical interest information of the target object;

[0014] Constructing a data feature set according to the historical travel information and the historical interest information;

[0015] The data feature set is standardized to obtain a standardized data feature set, and the standardized data feature set is analyzed based on cluster analysis to obtain the object attribute information.

[0016] Furthermore, the method for constructing the object attribute information further includes:

[0017] receiving an object attribute information establishment request sent by the navigation terminal; the object attribute information establishment request carries preset object attribute information; the object attribute information establishment request is obtained by operating the target object on the navigation terminal;

[0018] The preset object attribute information is subjected to object attribute information construction processing to obtain the object attribute information.

[0019] Furthermore, analyzing the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object includes:

[0020] Filtering the historical travel data according to the object attribute information to determine the historical travel data that matches the destination information;

[0021] filtering the historical interest information according to the object attribute information to determine the historical interest information that matches the destination information;

[0022] Acquire real-time situation information of the target object; the real-time situation information includes schedule information and weather information;

[0023] Based on the historical travel data matching the destination information, the historical interest information matching the destination information, and the real-time context information, the purpose information is analyzed for travel intention to obtain the travel intention.

[0024] Further, based on the travel intention, performing destination recommendation processing on the destination information, generating candidate recommended destinations, and sending the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations includes:

[0025] Based on the travel intention, determining a first candidate recommended destination within a preset distance of the destination information, and determining a second candidate recommended destination between the current location and the destination information;

[0026] determining the first candidate recommended destination and the second candidate recommended destination as the candidate recommended destinations;

[0027] sorting the candidate recommended destinations based on the distances between the candidate recommended destinations and the destination information to obtain a candidate recommended destination sequence;

[0028] The candidate recommended destination sequence is sent to the navigation terminal, so that the navigation terminal displays the candidate recommended destination sequence.

[0029] Furthermore, the sending the candidate recommended destination sequence to the navigation terminal so that the navigation terminal displays the candidate recommended destination sequence includes:

[0030] Determining, based on the object attribute information and the travel intention, a recommendation reason corresponding to each of the candidate recommended destinations;

[0031] The candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations are sent to the navigation terminal, so that the navigation terminal displays the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations.

[0032] Further, the performing route recommendation processing on the current navigation destination and the current location based on the object attribute information and the object navigation preference model, generating candidate recommended routes, and sending the candidate recommended routes to the navigation terminal so that the navigation terminal displays the candidate recommended routes, includes:

[0033] Based on the object navigation preference model, the real-time scenario information and the object attribute information, performing route recommendation processing on the current navigation destination and the current location, generating a preset candidate recommended route and a route description corresponding to each of the preset candidate recommended routes;

[0034] Based on the travel intention and the current navigation destination, determining recommended locations along the way corresponding to each of the preset candidate recommended routes;

[0035] constructing the candidate recommended routes based on the preset candidate recommended routes, the recommended places along the way corresponding to the preset candidate recommended routes, and the route descriptions corresponding to the preset candidate recommended routes;

[0036] Determining the priority corresponding to the object data information;

[0037] Sorting the candidate recommended routes according to the priorities corresponding to the object attribute information to obtain a recommended order of the candidate recommended routes;

[0038] According to the recommended order of the candidate recommended routes, the candidate recommended routes are sent to the navigation terminal, so that the navigation terminal displays the candidate recommended routes according to the recommended order of the candidate recommended routes.

[0039] Further, the determining the priority corresponding to the object data information includes:

[0040] Based on the historical travel data and the historical third-party software data, the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

[0041] Furthermore, the determining the priority corresponding to the object data information further includes:

[0042] A priority sorting request for the object attribute information sent by the navigation terminal is received, and the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

[0043] Furthermore, after sending the candidate recommended routes to the navigation terminal in the recommended order of the candidate recommended routes so that the navigation terminal displays the candidate recommended routes in the recommended order of the candidate recommended routes, the method further includes:

[0044] Acquire a current navigation route sent by the navigation terminal; the current navigation route is obtained by the target object selecting the candidate recommended route on the navigation terminal; the current navigation route carries current navigation type information;

[0045] When the current navigation type information and / or the real-time context information is updated, obtaining a new current position of the target object, new current navigation type information and new real-time context information;

[0046] Based on the new real-time situation information, the new current navigation type information and the travel intention, perform destination recommendation processing on the new current location and the current navigation destination, generate a new candidate recommended destination, and send the new candidate recommended destination to the navigation terminal so that the navigation terminal displays the new candidate recommended destination;

[0047] Acquire a new current navigation destination sent by the navigation terminal; the new current navigation destination is obtained by selecting the new candidate recommended destination by the target object on the navigation terminal;

[0048] Based on the object attribute information and the object navigation preference model, a route recommendation process is performed on the new current navigation destination to generate a new candidate recommended route, and the new candidate recommended route is sent to the navigation terminal so that the navigation terminal displays the new candidate recommended route.

[0049] Furthermore, the method for establishing the object navigation preference model includes:

[0050] The historical travel data is cleaned and converted into model data to obtain model data; the model data includes a training set and a test set;

[0051] Performing feature selection processing on the training set to determine model building data;

[0052] Performing split feature processing and division threshold processing on the model building data to build a first preset object navigation preference model;

[0053] Pruning the first preset object navigation preference model to obtain a second preset object navigation preference model;

[0054] Based on the test set, the second preset object navigation preference model is adjusted to obtain the object navigation preference model.

[0055] On the other hand, an embodiment of the present application further provides a navigation recommendation device, the device comprising:

[0056] A first acquisition module is used to obtain historical travel data of the target object, the current location of the target object and object attribute information in response to a navigation request of the target object; the navigation request carries destination information;

[0057] A determination module, configured to analyze the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object;

[0058] A first generating module is used to perform destination recommendation processing on the destination information based on the travel intention, generate candidate recommended destinations, and send the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations;

[0059] A second acquisition module is used to acquire a current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal;

[0060] The second generation module is used to perform route recommendation processing on the current navigation destination and the current location based on the object attribute information and the object navigation preference model, generate candidate recommended routes, and send the candidate recommended routes to the navigation terminal so that the navigation terminal displays the candidate recommended routes; the object navigation preference model is obtained by processing the historical travel data.

[0061] Furthermore, the device includes a construction module for object attribute information, and the construction module for object attribute information includes:

[0062] A historical third-party software data acquisition unit, used to acquire historical third-party software data of the target object; the historical third-party software data represents historical data of third-party software associated with the navigation terminal;

[0063] a historical travel information determination unit, configured to perform cleaning processing on the historical travel data to remove invalid data in the historical travel data and obtain historical travel information;

[0064] A historical interest information determination unit, configured to classify and filter the historical third-party software data to determine the historical interest information of the target object;

[0065] A data feature set construction unit, configured to construct a data feature set according to the historical travel information and the historical interest information;

[0066] The object attribute information construction unit is used to perform standardization processing on the data feature set to obtain a standardized data feature set, and analyze the standardized data feature set based on cluster analysis to obtain the object attribute information.

[0067] Furthermore, the object attribute information construction module is further used to receive an object attribute information establishment request sent by the navigation terminal; the object attribute information establishment request carries preset object attribute information; the object attribute information establishment request is obtained by operating the target object on the navigation terminal;

[0068] It is also used to perform object attribute information construction processing on the preset object attribute information to obtain the object attribute information.

[0069] Further, the determination module includes:

[0070] A first determining unit, configured to filter the historical travel data according to the object attribute information to determine the historical travel data that matches the destination information;

[0071] A second determining unit, configured to filter the historical interest information according to the object attribute information to determine the historical interest information that matches the destination information;

[0072] A real-time situation information acquisition unit, used to acquire the real-time situation information of the target object; the real-time situation information includes schedule information and weather information;

[0073] A travel intention analysis unit is used to perform travel intention analysis on the destination information based on the historical travel data matching the destination information, the historical interest information matching the destination information, and the real-time context information to obtain the travel intention.

[0074] Furthermore, the first generation module includes:

[0075] a first candidate recommended destination determining unit, configured to determine, based on the travel intention, a first candidate recommended destination within a preset distance of the destination information, and determine a second candidate recommended destination between the current location and the destination information;

[0076] a second candidate recommendation destination determining unit, configured to determine the first candidate recommendation destination and the second candidate recommendation destination as the candidate recommendation destinations;

[0077] a candidate recommended destination sequence acquisition unit, configured to sort the candidate recommended destinations based on the distances between the candidate recommended destinations and the destination information to obtain a candidate recommended destination sequence;

[0078] The candidate recommended destination sequence sending unit is used to send the candidate recommended destination sequence to the navigation terminal, so that the navigation terminal can display the candidate recommended destination sequence.

[0079] Furthermore, the candidate recommended destination sequence sending unit includes:

[0080] a recommendation reason determination unit, configured to determine a recommendation reason corresponding to each of the candidate recommended destinations based on the object attribute information and the travel intention;

[0081] The recommendation reason sending unit is used to send the candidate recommendation destinations and the recommendation reasons corresponding to the candidate recommendation destinations to the navigation terminal, so that the navigation terminal can display the candidate recommendation destinations and the recommendation reasons corresponding to the candidate recommendation destinations.

[0082] Furthermore, the second generation module includes:

[0083] a route description generating unit, configured to perform route recommendation processing on the current navigation destination and the current location based on the object navigation preference model, the real-time scenario information and the object attribute information, and generate a preset candidate recommended route and a route description corresponding to each of the preset candidate recommended routes;

[0084] A recommended location determination unit for determining the recommended locations along the way corresponding to each of the preset candidate recommended routes based on the travel intention and the current navigation destination;

[0085] A candidate recommended route construction unit, configured to construct the candidate recommended route based on the preset candidate recommended route, the recommended places along the route corresponding to each of the preset candidate recommended routes, and the route description corresponding to each of the preset candidate recommended routes;

[0086] A priority determination unit, used to determine the priority corresponding to the object data information;

[0087] A recommended order generating unit for candidate recommended routes, used to sort the candidate recommended routes according to the priorities corresponding to the object attribute information to obtain a recommended order for the candidate recommended routes;

[0088] The candidate recommended route sending unit is further used to send the candidate recommended routes to the navigation terminal according to the recommended order of the candidate recommended routes, so that the navigation terminal displays the candidate recommended routes according to the recommended order of the candidate recommended routes.

[0089] Furthermore, the priority determination unit is also used to prioritize the object attribute information based on the historical travel data and the historical third-party software data to obtain the priority corresponding to the object attribute information.

[0090] Furthermore, the priority determination unit is further configured to receive a priority sorting request for the object attribute information sent by the navigation terminal, perform priority sorting on the object attribute information, and obtain the priority corresponding to the object attribute information.

[0091] Furthermore, the device further comprises an update recommendation module, wherein the update recommendation module comprises:

[0092] A first acquisition unit is used to acquire a current navigation route sent by the navigation terminal; the current navigation route is obtained by the target object selecting the candidate recommended route on the navigation terminal; the current navigation route carries current navigation type information;

[0093] A second acquisition unit, configured to acquire a new current position, new current navigation type information and new real-time scenario information of the target object when the current navigation type information and / or the real-time scenario information is updated;

[0094] a destination recommendation unit, configured to perform destination recommendation processing on the new current location and the current navigation destination based on the new real-time scenario information, the new current navigation type information and the travel intention, generate a new candidate recommended destination, and send the new candidate recommended destination to the navigation terminal so that the navigation terminal displays the new candidate recommended destination;

[0095] A third acquisition unit is used to acquire a new current navigation destination sent by the navigation terminal; the new current navigation destination is obtained by the target object selecting the new candidate recommended destination on the navigation terminal;

[0096] A route recommendation unit is used to perform route recommendation processing on the new current navigation destination based on the object attribute information and the object navigation preference model, generate a new candidate recommended route, and send the new candidate recommended route to the navigation terminal so that the navigation terminal can display the new candidate recommended route.

[0097] Furthermore, the device further comprises an object navigation preference model construction module, and the object navigation preference model construction module comprises:

[0098] A model data acquisition unit, used for performing cleaning processing and model data conversion processing on the historical travel data to obtain model data; the model data includes a training set and a test set;

[0099] A model building data determination unit, used to perform feature selection processing on the training set to determine model building data;

[0100] A first preset model building unit, used for performing split feature processing and division threshold processing on the model building data to build a first preset object navigation preference model;

[0101] A second preset model building unit, configured to perform pruning processing on the first preset object navigation preference model to obtain a second preset object navigation preference model;

[0102] The object navigation preference model construction unit is used to adjust the second preset object navigation preference model based on the test set to obtain the object navigation preference model.

[0103] On the other hand, an embodiment of the present application also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the navigation recommendation method as described above.

[0104] On the other hand, an embodiment of the present application further provides a computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the navigation recommendation method as described above.

[0105] On the other hand, an embodiment of the present application further provides a computer program product, which implements the above-mentioned navigation recommendation method when the computer program is executed by a processor.

[0106] The embodiment of the present application provides a navigation recommendation method and device, the method comprising: in response to a navigation request of a target object, obtaining historical travel data of the target object, the current location of the target object and object attribute information; the navigation request carries destination information; according to the historical travel data and the object attribute information, the destination information is analyzed to determine the travel intention of the target object; based on the travel intention, destination recommendation processing is performed on the destination information to generate candidate recommended destinations, and the candidate recommended destinations are sent to a navigation terminal so that the navigation terminal displays the candidate recommended destinations; the current navigation destination sent by the navigation terminal is obtained; based on the object attribute information and the object navigation preference model, route recommendation processing is performed on the current navigation destination and the current location to generate a recommended destination. Candidate recommended routes are generated and sent to the navigation terminal so that the navigation terminal can display the candidate recommended routes. Therefore, when the user is navigating, the travel intention can be analyzed according to the historical travel data and object attribute information, so that the relevant places of the place information input by the user can be recommended to the user, and the navigation route can be recommended according to the object attribute information and the object navigation preference model. This not only realizes the in-depth mining of user navigation habits, generates travel routes that meet user preferences, and improves the navigation experience; it also realizes the accurate grasp of user navigation habits and real-time location information, provides users with comprehensive and personalized travel service recommendations, significantly improves the practicality and user satisfaction of navigation software, and has broad application prospects and market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0108] Figure 1 It is a schematic diagram of an implementation environment of a navigation recommendation method provided in an embodiment of the present application.

[0109] Figure 2 It is a flowchart of a navigation recommendation method proposed in an embodiment of the present application.

[0110] Figure 3 It is a flowchart of a method for constructing object attribute information provided in an embodiment of the present application.

[0111] Figure 4 It is a flowchart of another method for constructing object attribute information provided in an embodiment of the present application.

[0112] Figure 5 It is a flow chart of a method for determining travel intention provided in an embodiment of the present application.

[0113] Figure 6 It is a flowchart of a destination recommendation method provided in an embodiment of the present application.

[0114] Figure 7 It is a flowchart of a method for displaying candidate recommended destinations provided in an embodiment of the present application.

[0115] Figure 8 It is a flowchart of a method for determining a candidate recommended route provided in an embodiment of the present application.

[0116] Fig. 9 It is a flowchart of a method for updating candidate recommended routes provided in an embodiment of the present application.

[0117] Fig.10 It is a flowchart of a method for establishing an object navigation preference model provided in an embodiment of the present application.

[0118] Fig.11 It is a schematic diagram of a navigation recommendation device provided in an embodiment of the present application.

[0119] Fig.12 This is a hardware structure block diagram of a server of a navigation recommendation method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0120] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0121] It should be noted that the terms "first", "second", etc. in the specification and claims of the embodiment of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the embodiment of the present application described here can be implemented in an order other than those illustrated or described here. Thus, the features defined as "first" and "second" can expressly or implicitly include one or more of the features. In the description of the present embodiment, unless otherwise specified, the meaning of "multiple" is two or more. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0122] Figure 1 FIG. 1 is a schematic diagram of an implementation environment of a navigation recommendation method provided in an embodiment of the present application. Figure 1 As shown, the implementation environment may include at least a navigation terminal 01 and a server 02. The client 01 and the server 02 may be directly or indirectly connected via wired or wireless communication, which is not limited in this application.

[0123] Specifically, the navigation terminal 01 is a carrier of navigation software, and can be used to send a navigation request to the server 02. Optionally, the navigation terminal 01 can be a smart phone, a car navigation system, a portable navigation device, a smart watch, a tablet computer, an AR navigation device, a bicycle navigation device, and other devices with navigation functions. The embodiments of the present application are applied to various navigation scenarios.

[0124] Specifically, the server 02 can be used to respond to a navigation request of a target object, obtain historical travel data of the target object, the current location of the target object and object attribute information; the navigation request carries destination information; based on the historical travel data and object attribute information, the destination information is analyzed to determine the travel intention of the target object; based on the travel intention, the destination information is recommended to generate a candidate recommended destination, and the candidate recommended destination is sent to the navigation terminal so that the navigation terminal displays the candidate recommended destination; the current navigation destination sent by the navigation terminal is obtained; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal; based on the object attribute information and the object navigation preference model, the current navigation destination and the current location are recommended to generate a candidate recommended route, and the candidate recommended route is sent to the navigation terminal so that the navigation terminal displays the candidate recommended route; the object navigation preference model is obtained by processing the historical travel data. Optionally, the server 02 is a navigation server, which can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The server 02 can also be navigation software connected to the navigation server.

[0125] It should be noted that Figure 1 This is just an example. In other scenarios, other implementation environments may also be included.

[0126] Figure 2 It is a flowchart of a navigation recommendation method proposed in an embodiment of the present application. This specification provides method operation steps such as the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only execution order. When the actual system or the fourth server product is executed, it can be executed in sequence or in parallel according to the method shown in the embodiment or the accompanying drawings (for example, a parallel processor or a multi-threaded processing environment). Specifically, Figure 2 As shown, the method provided in the embodiment of the present application can be applied to navigation software, and the method may include:

[0127] S10: In response to a navigation request of the target object, historical travel data of the target object, the current location of the target object, and object attribute information are obtained; the navigation request carries destination information.

[0128] In the embodiment of the present application, the target object is a user using a navigation terminal, and a navigation request refers to a navigation instruction input into the navigation terminal, which may be a voice instruction or a text instruction; the navigation request carries destination information, and the destination information refers to detailed information of the final location or target location specified by the user when navigating. Object attribute information refers to user tags or user portraits, which refer to detailed descriptions constructed based on user behavior, preferences, background, and other information. The historical travel data of the target object refers to the travel data of the target object on each navigation terminal before this navigation. As long as the target object logs in to the same account on each navigation terminal, these historical travel data can be obtained.

[0129] Responding to the navigation instruction of the target object refers to analyzing and identifying the navigation instruction to determine the destination information or location information corresponding to the navigation instruction. Optionally, when the navigation instruction is a voice instruction, the voice is converted into text and location features are extracted; when the navigation instruction is a text instruction, location features are extracted from the text instruction.

[0130] After responding to the navigation instruction of the target object, the historical travel data of the target object, the current location of the target object and the object attribute information are obtained to provide basic data for subsequent navigation recommendations.

[0131] Optionally, navigation instructions include but are not limited to navigating to a certain place, searching the surrounding area, querying a location, adding to favorites, adding waypoints, querying road status, and querying road speed limits.

[0132] Optionally, the destination information includes but is not limited to the specific address of the destination (including street name, house number, etc.), the name of the destination (such as a store, restaurant, attraction, company, etc.), geographic coordinates (latitude and longitude coordinates of the destination, used for precise positioning), the type or category of the destination (such as catering, entertainment, shopping, accommodation, etc.), contact information (contact number or official website link of the destination), opening hours (business hours or opening hours of the destination), user evaluation (other users' evaluations and comments on the destination), pictures (pictures of the exterior or interior of the destination), transportation methods (recommended transportation methods to reach the destination, such as walking, driving, public transportation, etc.), surrounding facilities (information on facilities around the destination, such as parking lots, toilets, ATM machines, etc.), navigation routes (route guidance from the current location to the destination) or more.

[0133] Optionally, object attribute information is usually displayed in the form of a label, such as travel expert, fisherman, office worker, running enthusiast, food lover, etc., which can display the user's interests or navigation rules and route requirements.

[0134] Optionally, each target object can have multiple object attribute information.

[0135] Optionally, the current position of the target object may be acquired through positioning technology in the navigation terminal.

[0136] Optionally, the method for obtaining historical travel data may include the following steps:

[0137] (1) Determine data requirements: Identify the types of data that need to be collected. Optionally, this may include the user's starting and ending coordinates, waypoints, selected navigation route information, navigation category (such as walking, driving, cycling, etc.), departure time, arrival time at destination, whether to continue using navigation at destination, vehicle speed data during driving, facial expression images or videos during driving, weather conditions, schedule, etc.

[0138] (2) Determine the data source: Determine the data source corresponding to the data requirement. Optionally, the starting and ending points of the user's navigation, departure time, arrival time at the destination, whether to continue using navigation at the destination, the selected navigation route and navigation category can be obtained from the navigation software or the system's internal database; vehicle speed data can be obtained from the vehicle's OBD (On Board Diagnostics) interface or GPS (Global Positioning System) tracking device; walking data or cycling data can be obtained from sports and health software on mobile phones or smart wearable devices; facial expression images or videos during driving can be collected from the vehicle's camera or the wearable device worn by the driver; weather conditions and schedules can be obtained from the vehicle's system, mobile phone or smart wearable device.

[0139] (3) Collect data: Collect data from data sources according to data requirements. It should be noted that collecting data from third-party software or other terminals requires obtaining permission from these devices or software.

[0140] (4) Data cleaning: First, the collected data is cleaned to remove duplicate, invalid or erroneous data. Then, data is derived and the surrounding areas of the destination are radiated. The scenic spots, gas stations, libraries, fishing spots, etc. around each navigation destination are stored accordingly.

[0141] Through the above steps, the historical travel data is obtained.

[0142] It should be noted that in the specific implementation of this application, user location-related data is involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0143] S30: Analyze the destination information based on the historical travel data and the object attribute information to determine the travel intention of the target object.

[0144] In an embodiment of the present application, when navigating, the destination information is first analyzed based on the user's historical travel data and object attribute information, so as to filter out the content in the historical travel data and / or the object attribute information that matches or is related to the destination information, and based on these related contents, the travel intention of the target object is determined. The travel intention can be the reason or purpose of the target object's navigation, such as going to work, traveling, fishing, shopping, etc. For example, when the destination information of the target object is a place near the river, and the user navigates to the river more than a preset number of times, and the user has the object attribute information of "Fisherman", it can be inferred that the travel intention of the target object is fishing.

[0145] S50: Based on the travel intention, perform destination recommendation processing on the destination information, generate candidate recommended destinations, and send the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations.

[0146] In the embodiment of the present application, the destination recommendation process is to calculate and recommend the destination of this navigation according to the travel intention and the destination information, thereby obtaining a candidate recommended destination, and sending the candidate recommended destination to the navigation terminal, so that the navigation terminal can display the candidate recommended destination. The candidate destination includes multiple recommended destinations obtained based on the above method. The navigation terminal displays the candidate recommended destination on the human-computer interaction interface of the navigation terminal, specifically, the candidate recommended destination is displayed to the target object on the navigation software interface.

[0147] Optionally, when the destination information carries specific address information, or other information that can represent the specific address, recommendations are made based on the travel intention and the specific address information. For example, if the travel intention is fishing, and the specific address information is a location near a river, then fishing spots near the river and fishing spots between the current location and the river are recommended, which together constitute candidate recommended destinations.

[0148] S70: Acquire the current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting candidate recommended destinations on the navigation terminal.

[0149] In an embodiment of the present application, after the candidate recommended destinations are sent to the navigation terminal, the current navigation destination sent by the navigation terminal is obtained. The current navigation destination refers to the destination of this navigation, which is the destination selected by the target object from the candidate recommended destinations in the navigation terminal, that is, the current navigation destination is one of the candidate recommended destinations.

[0150] S90: Based on the object attribute information and the object navigation preference model, route recommendation processing is performed on the current navigation destination and the current location, candidate recommended routes are generated, and the candidate recommended routes are sent to the navigation terminal so that the navigation terminal displays the candidate recommended routes; the object navigation preference model is obtained by processing historical travel data.

[0151] In the embodiment of the present application, before navigation is performed, the server processes the historical travel data to obtain the object navigation preference model. Route recommendation processing refers to analyzing the current navigation destination and the current location according to the object attribute information and the object navigation preference model, thereby performing route planning, and obtaining multiple recommended routes that meet the object navigation preference and the object attribute information. The multiple recommended routes that meet the object navigation preference and the object attribute information constitute candidate recommended routes.

[0152] For example, if the object attribute information is "fisherman" and "travel expert", the current navigation destination is a fishing spot, and the user's navigation preferences are the closest distance and the best scenery along the way, then when making route recommendations, recommendations will be made based on this information, such as "the closest route", "the best scenery along the way", "the route with the closest distance and the most fishing spots along the way", "the route with the most fishing spots along the way", "the route with a fishing tackle shop along the way", etc., and these routes will be sent to the navigation terminal as candidate recommended routes, so that the navigation terminal can display the candidate recommended routes and the target object can choose according to its own needs. The above embodiment introduces the situation when the target object has multiple object attribute information and multiple navigation preferences. When the target object has only one object attribute information and one navigation preference, the recommendation is mainly based on the one object attribute information and one navigation preference.

[0153] Optionally, when the candidate recommended routes are displayed on the navigation terminal, they are accompanied by route descriptions related to the object's navigation preferences and / or object attribute information, so that the target object can clearly understand the candidate recommended routes.

[0154] Optionally, when the candidate recommended routes are displayed on the navigation terminal, there are recommendations of places along the way related to the object attributes. For example, when displaying the "route with the most fishing spots along the way", information such as fishing tackle shops and convenience stores related to fishing will be marked or introduced on the navigation terminal at the same time.

[0155] Optionally, when the candidate recommended routes are displayed on the navigation terminal, they are accompanied by recommendations of places along the way related to living facilities, such as restaurants, parking, etc.

[0156] Optionally, high-risk driving areas, historical violation points of users, scenic spots, parking lots, restaurants, etc. in the candidate recommended routes are marked with different colors. When the navigation software passes through the recommended destination or recommended area calculated by the object navigation preference model and object attribute information during the user's use, it will be highlighted, and the user can see the corresponding mark of the recommended destination or recommended area in the overall route view.

[0157] Optionally, in the above steps S10-S90, the display method on the navigation terminal includes but is not limited to one or more of display on the human-computer interaction interface, voice broadcast, and integration of smart wearable devices. Specifically, it can include map display, voice navigation, text prompts, icons and symbols, real-time traffic information, 3D format, street view, offline map, augmented reality (AR) navigation, etc.

[0158] In the embodiment of the present application, when a navigation request is received, the current location, historical travel data and object attributes of the target object are obtained, which can be used to analyze the destination information carried in the navigation request, thereby obtaining the user's travel intention, so that the destination can be recommended to the target object according to the travel intention, and personalized recommendations for navigation destinations can be made according to user portraits. After obtaining the current navigation destination, route planning is performed according to the user portrait and user navigation preferences, thereby generating multiple personalized candidate recommended routes for users to choose from, which not only realizes in-depth mining of user navigation habits, generates travel routes that meet user preferences, and improves navigation experience; it also realizes accurate grasp of user navigation habits and real-time location information, and provides users with all-round and personalized travel service recommendations, significantly improving the practicality and user satisfaction of navigation software, and has broad application prospects and market value.

[0159] As an optional implementation, Figure 3 is a flow chart of a method for constructing object attribute information provided in an embodiment of the present application, such as Figure 3 As shown, the method for constructing object attribute information includes:

[0160] S001: Acquire historical third-party software data of a target object; the historical third-party software data represents historical data of third-party software associated with a navigation terminal.

[0161] In the embodiment of the present application, when constructing object attribute information, it is necessary to first obtain historical travel data and historical third-party software data. The method for obtaining historical travel data is the same as the method in step S10, which will not be repeated here.

[0162] In the embodiment of the present application, the object attribute information is obtained based on the historical travel data and the historical third-party software data, and the historical third-party software data represents the historical data of the third-party software associated with the navigation terminal, wherein the third-party software refers to other software on the navigation terminal, including but not limited to entertainment software, shopping software, travel software, news software or other navigation software, etc. After authorization, the data of the third-party software can be obtained by the navigation software to which the present method is applied.

[0163] S003: Clean the historical travel data to remove invalid data in the historical travel data and obtain historical travel information.

[0164] In the embodiment of the present application, the historical travel data obtained in the above steps is cleaned again. Data cleaning is an important step in data preprocessing, which aims to improve data quality and ensure the accuracy of data analysis and modeling.

[0165] Specifically, cleaning can include identifying and handling missing values, as well as identifying and handling outliers:

[0166] Outliers are data points that deviate significantly from other values ​​in the data set. Optionally, the method for removing outliers can be the following methods. Statistical analysis method: use the 3σ principle (that is, data values ​​outside three standard deviations of the mean are considered outliers) or box plot analysis (based on quartiles and interquartile ranges, usually points outside 1.5 times the IQR are considered outliers); data smoothing: smooth the data through binning or regression analysis to reduce the impact of outliers; direct deletion: if the outliers are caused by errors, the records containing outliers can be directly deleted; treat as missing values: treat outliers as missing values, and then use the missing value processing method to handle them.

[0167] Optionally, the missing values ​​can be handled by deletion, filling or no processing. Specifically, delete tuples: if the object has multiple attribute missing values, you can choose to delete these records, but this requires that the proportion of objects with missing values ​​is small; data filling: fill the empty values ​​with certain values, usually based on statistical principles, such as using the mean, median or mode to fill; interpolation filling: use interpolation methods, such as forward filling (filling with the previous non-empty value), backward filling (filling with the next non-empty value) or linear interpolation.

[0168] S005: Classify and filter historical third-party software data to determine historical interest information of the target object.

[0169] In an embodiment of the present application, by classifying and screening historical third-party software data, the historical interest information of the target object can be determined. The historical interest information refers to the interests, preferences, hobbies or areas of concern shown by the target object in the past period of time.

[0170] Optionally, historical interest information is mainly used to analyze and screen third-party software data from the following aspects: browsing history (pages and content that users have browsed on websites or applications, which can reflect their interests), search history (keywords entered by users in search engines or platforms, which reveal the information or products they are looking for), purchase history, social media interactions (content that users like, comment on and share on social media, which reflects their personal interests and preferences in social circles), subscriptions and collections (newsletters, blogs, channels or favorite products subscribed by users, which are their areas of interest), application usage (the frequency and duration of users' use of mobile applications, and their behavior in applications, such as game progress, reading habits, etc.), geographic location information (user location data, including places they frequently visit, which can reveal their living habits and possible interests), feedback and evaluation (user evaluation and feedback on products or services, which can provide direct evidence of user satisfaction and preferences), and questionnaires and market research (questionnaires and market research in which users participate, which can provide detailed views and preferences of users on specific topics).

[0171] Optionally, the past period of time is not specifically limited in this application and may be the past week, the past month, the past year, or a longer or shorter period of time.

[0172] S007: Construct a data feature set based on historical travel information and historical interest information;

[0173] In the embodiment of the present application, after obtaining the historical travel information, a data feature set is constructed for the historical travel information and the historical interest information. The construction method may be to determine the weights corresponding to the historical travel information and the historical interest information, and construct the data feature set based on the weights corresponding to the historical travel information and the historical interest information.

[0174] S009: Standardize the data feature set to obtain a standardized data feature set, and analyze the standardized data feature set based on cluster analysis to obtain object attribute information.

[0175] In the embodiment of the present application, the data feature set is standardized, that is, format conversion is performed, and the data is converted into a format suitable for cluster analysis, such as a feature vector, to obtain a standardized data feature set.

[0176] After obtaining the standardized data feature set, you need to select the type of clustering algorithm. Specifically, select a suitable clustering algorithm according to the characteristics of the standardized data feature set and the analysis requirements. Optionally, you can use affinity propagation (AffinityPropagation), DBSCAN (Density-Based Spatial Clustering of Applications withNoise, density-based spatial clustering applications with noise), K-means (K-means clustering algorithm), K-medoids (K-medoids clustering algorithm), Mean Shift (mean shift clustering algorithm), OPTICS (sorting points to determine cluster structure), Gaussian mixture model (GMM), spectral clustering, Blondel community partitioning algorithm, etc. for analysis. The following takes K-means as an example to analyze the standardized data feature set based on cluster analysis and introduce the object attribute information obtained:

[0177] (1) Initialization: Randomly select K data points in the standardized data feature set as the initial cluster centers.

[0178] (2) Assignment: Assign each data point in the standardized data feature set to the nearest cluster center.

[0179] (3) Update: Recalculate the center of each cluster (i.e., the mean of all points in the cluster).

[0180] (4) Iteration: Repeat steps (2) and (3) until the cluster center no longer changes or the preset number of iterations is reached to obtain the clustering result.

[0181] (5) Based on the clustering results, the target objects are divided into different groups, and the common characteristics and behavior patterns of each group are analyzed. Based on the common characteristics and behavior patterns of each group, object attribute information is established for each group, such as "travel maniac", "fishing man", "city commuter", etc.

[0182] It should be noted that steps S001-S009 are applicable when the number of historical travel data obtained exceeds a preset number, that is, only when the number of historical travel data exceeds a preset number can the object attribute information be constructed.

[0183] The embodiment of the present application analyzes the historical travel data and third-party software data of the target object to analyze the user behavior patterns, thereby performing clustering and obtaining the object attribute information of the target object, which is conducive to realizing personalized recommendations based on the target object, improving the navigation experience, and thus improving the practicality of the navigation software and user satisfaction, and has broad application prospects and market value.

[0184] As an optional implementation, Figure 4 is a flow chart of another method for constructing object attribute information provided in an embodiment of the present application. Figure 4 As shown, the method for constructing object attribute information may also include:

[0185] S0001: receiving an object attribute information establishment request sent by a navigation terminal; the object attribute information establishment request carries preset object attribute information; the object attribute information establishment request is obtained by operating a target object on the navigation terminal;

[0186] S0003: Perform object attribute information construction processing on the preset object attribute information to obtain object attribute information.

[0187] In the embodiment of the present application, in addition to steps S001-S009 for establishing object attribute information, the target object can also establish object attribute information autonomously. The specific method includes:

[0188] The target object initiates an object attribute information establishment request in the navigation terminal, and the object attribute information establishment request carries preset object attribute information, which refers to the user tag or user portrait selected by the target object to be established, and the preset object attribute information can be pre-configured in the navigation software of the navigation terminal, or can be established by the user. In other words, the navigation software used in the method of the embodiment of the present application has the function of autonomously establishing object attribute information.

[0189] After receiving the object attribute information establishment request sent by the terminal device, the preset object attribute information carried therein is identified, the preset object attribute information is constructed, and the object attribute information is obtained.

[0190] Optionally, the object attribute information establishment request may carry one or more preset object attribute information, so the constructed object attribute information refers to more than one, but is a general term for multiple object attribute information.

[0191] The embodiment of the present application receives an object attribute information establishment request sent by a navigation terminal, constructs and processes the preset object attribute information, and enables the target object to independently establish the object attribute information, thereby accurately grasping the user's navigation habits, laying the foundation for realizing personalized travel recommendation services, thereby significantly improving the practicality and user satisfaction of the navigation software, and having broad application prospects and market value.

[0192] As an optional implementation, Figure 5 is a flow chart of a method for determining travel intention provided in an embodiment of the present application, such as Figure 5As shown, in the above step S30, the above step analyzes the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object, including:

[0193] S301: Filter and process the historical travel data according to the object attribute information to determine the historical travel data that matches the destination information.

[0194] In the embodiment of the present application, the object attribute information is taken as a premise, and the data matching the destination information is screened out in the historical itinerary data to obtain the historical itinerary data matching the destination information. For example, if the object attribute information is a travel expert and a fisherman, then when performing the screening process, the tourist attractions and fishing spots that the target object around the destination information has navigated are preferentially screened, and the places related to the tourist attractions and fishing spots around the destination are also screened, such as fishing tackle shops, scenic shopping centers, etc. Therefore, matching includes any one of exact matching, fuzzy matching, type matching, partial matching, geographic location matching, and user behavior matching. That is, any one of the above matching types is screened out as historical itinerary data matching the destination information.

[0195] Specifically, an exact match means that the name, address or geographic location (latitude and longitude) of the destination is exactly the same as the destination or waypoint in the historical itinerary data. A fuzzy match means that even if it is not exactly the same, the name or address of the destination is semantically similar or related to the information in the historical itinerary data. For example, "Central Park" and "Downtown Park" may be considered a fuzzy match. A geographic location match means that the geographic location (latitude and longitude) of the destination is similar to the geographic location in the historical itinerary data within a certain error range, which usually involves geocoding and reverse geocoding technology in the Geographic Information System (GIS). A partial match means that the destination information is part of the destination in the historical itinerary data. For example, if the destination in the historical itinerary data is "a street in Chaoyang District, Beijing", and the filter condition is "Chaoyang District, Beijing", they are considered to be partial matches. A type match means that the destination belongs to a certain category, and the destination in the historical itinerary data also belongs to the same category. For example, if the destination is "restaurant", then all the data in the historical itinerary with the destination of various restaurants as the destination are considered to be matched. A user behavior match means that the user's behavior at the destination matches the behavior recorded in the historical itinerary data, such as the user's stay time at the destination, the frequency of visits, etc.

[0196] Optionally, when making navigation recommendations, priorities can be assigned to the above-mentioned matching types. For example, if this method focuses more on personalized recommendations, then the priority of user behavior matching will be adjusted to first, so that the matched data will focus more on matching user behavior. The specific priority sorting method can be set by the target user through the navigation terminal, or the navigation server can sort based on the amount of matched historical travel data. For example, the amount of historical travel data that accurately matches the destination information is the largest, so the accurate match is given the highest priority, thereby establishing a matching type priority list. The next time you match, you can match based on the matching type priority list. After the matching is completed, the matching type priority list is updated based on the matching results to obtain an updated matching type priority list, which is then used as a cycle.

[0197] S303: Filter the historical interest information according to the object attribute information to determine the historical interest information that matches the destination information.

[0198] In the embodiment of the present application, according to the object attribute information, the historical interest information is screened and processed to screen out the historical interest information that matches the destination information. For example, if the object attribute information is a fisherman, then when the screening process is performed, the type of the destination information is determined. Assuming that the destination information is near the river, then the third-party software data is screened to determine whether the user has browsed the fishing spot information near the river, whether the third-party fishing software (such as Diaoyubao) has been opened, and whether the navigation is opened through the third-party fishing software (such as Diaoyubao), etc. Based on this information, it can be determined that the user interest information is fishing. In the historical interest information, the data that matches the destination information is screened out to obtain the historical interest information that matches the destination information. Here, matching refers to any one of exact matching, fuzzy matching, type matching, partial matching, geographic location matching, and user behavior matching. That is, any one of the above matching types is screened as the historical itinerary data that matches the destination information. The above matching method and matching type priority sorting method are the same as S301, which will not be repeated here.

[0199] S305: Acquire real-time situation information of the target object; the real-time situation information includes schedule information and weather information.

[0200] In the embodiments of the present application, the acquisition of real-time context information refers to the ability to use a variety of sensors and data analysis technologies to understand the user's specific environment and needs, and provide customized services and responses accordingly. Real-time context information includes schedule information and weather information, as well as the user's mood, the user's current location, and traffic conditions around the user's current location.

[0201] S307: Based on the historical travel data matching the destination information, the historical interest information matching the destination information, and the real-time context information, the destination information is analyzed for travel intention to obtain the travel intention.

[0202] In the embodiment of the present application, the travel intention of the target object can be analyzed according to the historical travel data matched with the destination information, the historical interest information matched with the destination information, and the real-time context information obtained in the above steps, so as to obtain the travel purpose of the target object. For example, the destination information is near the river, the historical travel data matched with the destination information is also near the river, and the historical interest information matched with the destination information is the user's first use of a third-party fishing software (such as Diaoyubao) to open the navigation, and the current user's schedule does not conflict with fishing in the current time period or within the time of arriving near the river, then the travel intention is analyzed to be fishing; the current user's schedule has a schedule that conflicts with fishing in the current time period or within the time of arriving near the river, and the travel intention can be determined according to the schedule, or a message prompt can be sent to the navigation terminal to prompt the target object that there is a schedule, whether to continue navigating according to the destination information or according to the schedule.

[0203] The embodiments of the present application can determine the user's travel intention by analyzing historical travel data, historical interest information and real-time situational information related to destination information, laying the foundation for personalized navigation recommendations, and can realize real-time perception of user location and situational changes, and provide timely and applicable travel and service recommendations, thereby accurately grasping the user's navigation habits and real-time location information, providing users with comprehensive and personalized travel service recommendations, significantly improving the practicality and user satisfaction of navigation software, and having broad application prospects and market value.

[0204] As an optional implementation, Figure 6 is a flowchart of a destination recommendation method provided by an embodiment of the present application, such as Figure 6 As shown, in the above step S50, based on the travel intention, the destination information is processed for destination recommendation, candidate recommended destinations are generated, and the candidate recommended destinations are sent to the navigation terminal so that the navigation terminal displays the candidate recommended destinations, including:

[0205] S501: Based on the travel intention, determine a first candidate recommended destination within a preset distance of the destination information, and determine a second candidate recommended destination between the current location and the destination information.

[0206] In the embodiment of the present application, places within the preset address range corresponding to the destination information and places related to the travel intention are screened out as the first candidate recommended destination. For example, if the travel intention is fishing, all fishing spots, fishing tackle shops and other places related to fishing within the preset address range corresponding to the destination information are screened out as the first candidate recommended destination.

[0207] And filter out places related to the travel intention between the address corresponding to the destination information and the current location as the second candidate recommended destination. For example, if the travel intention is fishing, then filter out all fishing spots, fishing tackle shops and other places related to fishing between the address corresponding to the destination information and the current location as the second candidate recommended destination.

[0208] Optionally, the preset range is 0-10 km around the destination information, and can also be set and changed by the target object on the navigation terminal.

[0209] Optionally, in addition to the recommended places related to the travel intention, the candidate destination recommendation also needs to mark and recommend the infrastructure around the candidate destination, such as parking lots, restaurants, convenience stores and other infrastructure, so that the target object can have a clearer understanding of the candidate recommended destination.

[0210] S503: Determine the first candidate recommended destination and the second candidate recommended destination as candidate recommended destinations.

[0211] In the embodiment of the present application, the set of the first candidate recommended destination and the second candidate recommended destination is determined as the candidate recommended destination.

[0212] It should be noted that the first candidate recommended destination, the second candidate recommended destination, and the candidate recommended destination, etc. in this specification are all collections of multiple destinations, and do not refer to a single location.

[0213] S505: Sort the candidate recommended destinations based on the distances between the candidate recommended destinations and the destination information to obtain a candidate recommended destination sequence.

[0214] In an embodiment of the present application, the candidate recommended destination sequence refers to a candidate destination list, and a method for determining the candidate recommended destination sequence includes: first determining the distance between each location in the candidate recommended destination and the location in the destination information, and sorting the candidate recommended destinations according to a preset distance rule.

[0215] Optionally, the preset distance rule refers to a sorting rule for candidate recommended destinations, which is not specifically limited in this application. The sorting rule may be to first display the candidate recommended destinations closest to the destination information, then display the candidate recommended destinations close to the current location, and finally display the candidate recommended destinations from near to far from the current location. Preferably, all candidate recommended destinations are displayed in a map format on the navigation interface of the navigation software.

[0216] S507: Sending the candidate recommended destination sequence to the navigation terminal, so that the navigation terminal displays the candidate recommended destination sequence.

[0217] In the implementation of the present application, after the candidate recommended destination sequence is obtained, it is sent to the navigation terminal, and the navigation terminal displays the candidate recommended destination sequence based on the human-computer interaction interface so that the target object can select the candidate recommended destination.

[0218] The embodiment of the present application defines a method for determining a candidate recommended destination, which can determine a first candidate recommended destination located at a preset distance from the destination information, and a second candidate recommended destination located between the current location and the destination information. The first candidate recommended destination and the second candidate recommended destination constitute a candidate recommended destination, and the candidate recommended destinations are sorted to obtain a candidate recommended destination sequence, so that the navigation terminal can display the candidate recommended destinations more clearly, thereby facilitating real-time perception of user location and context changes, and providing timely and applicable travel and service recommendations. It provides users with all-round and personalized travel service recommendations, significantly improves the practicality and user satisfaction of navigation software, and has broad application prospects and market value.

[0219] As an optional implementation, Figure 7 is a flowchart of a method for displaying candidate recommended destinations provided in an embodiment of the present application, such as Figure 7 As shown, in the above step S507, the sending of the candidate recommended destination sequence to the navigation terminal so that the navigation terminal displays the candidate recommended destination sequence includes:

[0220] S5071: Based on the object attribute information and the travel intention, determine the recommendation reasons corresponding to each candidate recommended destination.

[0221] S5073: Sending the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations to the navigation terminal, so that the navigation terminal displays the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations.

[0222] In the embodiment of the present application, the recommendation reasons corresponding to the candidate recommended destinations are determined according to the object attribute information and the travel intention, and the reasons for the respective recommendations can also be referred to as the location descriptions corresponding to the candidate recommended destinations. The location description includes but is not limited to the geographical location, surrounding environment, traffic conditions, functional use, weather conditions, facilities and services, flow of people, safety conditions, cultural characteristics, historical background, etc.

[0223] After determining the recommendation reasons corresponding to each candidate recommended destination, send them to the navigation terminal so that the navigation terminal can display the candidate recommended destinations and the recommendation reasons corresponding to each candidate recommended destination. Optionally, the display method includes one of voice and text or a combination thereof, which can be determined independently according to the user's choice. Generally, it is displayed based on the map software of the navigation terminal. For example, when the travel intention is fishing and the object attribute information is a fisherman, when the user enters the end point or waypoint, the location of multiple fishing spots in the surrounding area will be prompted, and the characteristics of the good fishing spots will be marked (whether it is a popular fishing spot, whether the fish situation is good or bad, whether it is convenient for parking, etc.).

[0224] In the embodiment of the present application, by determining the recommendation reasons corresponding to each candidate recommended destination and sending them to the client together with the candidate recommended destination, the target object can clearly understand each candidate recommended destination, which is conducive to providing timely and applicable travel and service recommendations. It is conducive to providing users with comprehensive and personalized travel service recommendations, significantly improving the practicality and user satisfaction of navigation software, and has broad application prospects and market value.

[0225] As an optional implementation, Figure 8 is a flow chart of a method for determining a candidate recommended route provided in an embodiment of the present application, such as Figure 8 As shown, in the above step S90, based on the object attribute information and the object navigation preference model, a route recommendation process is performed on the current navigation destination and the current location, a candidate recommended route is generated, and the candidate recommended route is sent to the navigation terminal so that the navigation terminal displays the candidate recommended route, including:

[0226] S901: Based on the object navigation preference model, real-time scenario information and object attribute information, route recommendation processing is performed on the current navigation destination and the current location, and preset candidate recommended routes and route descriptions corresponding to the preset candidate recommended routes are generated.

[0227] In the embodiment of the present application, a route description is generated for all preset candidate recommended routes. Specifically, the route recommendation process is generated by combining the object navigation preference model, real-time scenario information and object attribute information, and preset candidate recommended routes and corresponding route descriptions are generated at the same time. The route description can be generated based on one or more of the above-mentioned multiple evaluation objects.

[0228] As an optional implementation, before the above step S901, the method further includes: determining a navigation type recommendation priority according to a distance between a current position and a current navigation destination, and an object navigation preference model.

[0229] In the embodiment of the present application, when the navigation route is recommended, the navigation type recommendation is usually also included. In general, recommended routes are provided based on different navigation types. However, in order to improve the personalization of the recommendation, the embodiment of the present application determines the navigation type recommendation priority, which is specifically determined based on the distance between the current position and the current navigation destination, and the object navigation preference model. The object navigation preference model generates a model corresponding to the distance and the navigation type, so that the navigation type recommendation priority can be generated according to the distance. When the distance between the current distance and the current navigation destination is less than the first preset distance, the priority is ranked as travel mode A, travel mode B, travel mode C, travel mode D, etc.; when it is greater than the first preset distance and less than the second preset distance, the priority is ranked as travel mode B, travel mode A, travel mode C, travel mode D, etc.; when it is greater than the second preset distance and less than the third preset distance, the priority is ranked as travel mode C, travel mode B, travel mode A, travel mode D, etc. The above is only an example. The specific first preset distance, second preset distance and third preset distance are determined based on the object navigation preference model. Travel modes A, B, C and D are only examples and do not refer to specific modes. When making actual recommendations, they are also determined based on the object navigation bias model. Therefore, when making navigation route recommendations, the priority of navigation type recommendation can also be combined.

[0230] For example, when the object navigation preference model predicts a preference for tree-lined roads, the real-time scenario information indicates that the weather is clear and there are no daily arrangements, and the object attribute information is a travel expert or a walking enthusiast, and the current navigation destination is a scenic spot, the preset recommended route generated at this time may be one that satisfies the preference for tree-lined roads and beautiful scenery along the way, and the corresponding route description is "a scenic route"; it may also be one that satisfies the preference for tree-lined roads and many attractions along the way, and the corresponding route description is "many attractions and suitable for walking". The route description is mainly based on reflecting as many characteristics of the route as possible, and the route recommendation is mainly based on trying to meet the navigation type recommendation priority, object navigation preference model, real-time scenario information and object attribute information; when route planning cannot meet the above conditions at the same time, route recommendations are made based on one or more of the above conditions.

[0231] S902: Based on the travel intention and the current navigation destination, determine the recommended places along the way corresponding to each preset candidate recommended route.

[0232] In the embodiment of the present application, the user will be recommended places along the way according to the user's travel intention. The recommended places along the way corresponding to each preset candidate recommended route refer to places associated with the travel intention, or associated basic settings, and the concept of association is the same as above.

[0233] S903: constructing candidate recommended routes based on preset candidate recommended routes, recommended places along the routes corresponding to the preset candidate recommended routes, and route descriptions corresponding to the preset candidate recommended routes;

[0234] In an embodiment of the present application, the above-mentioned preset candidate recommended routes, the recommended places along the way corresponding to each preset candidate recommended route, and the route description corresponding to each preset candidate recommended route are constructed into candidate recommended routes, that is, the candidate recommended routes include route descriptions, place recommendations along the way, and route planning.

[0235] S904: Determine the priority corresponding to the object data information.

[0236] As an optional implementation, in the above step S904, the above determination of the priority corresponding to the object data information includes:

[0237] Based on historical travel data and historical third-party software data, the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

[0238] In an embodiment of the present application, the trip ratio corresponding to each object attribute information is obtained based on the number of trips corresponding to the object attribute information in the historical trip data, and the ratio of the number of trips for each object attribute information to the total number of trips in the historical trip data is calculated; and the amount of data corresponding to the object attribute information in the historical third-party software data is obtained based on the ratio of the amount of data corresponding to each tag to the total amount of data in the historical third-party software data to obtain the interest ratio corresponding to each object attribute information; the trip ratio and interest ratio corresponding to each trip are added, and the larger the value obtained by the addition, the higher the priority corresponding to the object attribute information.

[0239] Prioritize the object attribute information to obtain the priority corresponding to the object attribute information. For example, when the target object has three tags: "travel expert", "city commuter", and "fisherman", determine the number of trips related to the three tags in the historical trip data, and convert the number of trips into trip ratio data; and determine the amount of data related to the three tags in the historical third-party software data (including the number of visits, visit time, etc.), and convert it into interest ratio data; add up the two ratio data corresponding to each tag, and the larger the sum, the higher the priority.

[0240] As an optional implementation, in the above step S904, the above determination of the priority corresponding to the object data information further includes:

[0241] A priority sorting request for the object attribute information sent by the navigation terminal is received, and the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

[0242] In an embodiment of the present application, the method for prioritizing object attribute information can also be performed based on a sorting request of the target object, which is similar to the method for establishing the priority of the object attribute information described above, and is mainly based on autonomous establishment by the target object, and specifically includes: the target object initiates an object attribute information priority sorting request at the navigation terminal, and the existing object attribute information is prioritized based on the preset object attribute information priority carried in the object attribute information priority sorting request, so that the priority corresponding to the object attribute information can be obtained.

[0243] The embodiment of the present application uses a system to automatically prioritize object attribute information based on data analysis and prioritize object attribute information based on user requests, so that when making navigation recommendations, the priority of candidate recommended routes can be prioritized based on the priority of the object attribute information, which is conducive to achieving high personalization, that is, deeply mining user navigation habits, generating travel routes that meet user preferences, improving navigation experience, and providing users with comprehensive and personalized travel service recommendations, significantly improving the practicality of navigation software and user satisfaction, and has broad application prospects and market value.

[0244] S905: Sorting the candidate recommended routes according to the priorities corresponding to the object attribute information to obtain a recommendation order of the candidate recommended routes.

[0245] In an embodiment of the present application, the candidate recommended routes are sorted according to the priority corresponding to the acquired object attribute information, so that the recommended order of the candidate recommended routes can be obtained. Specifically, the candidate recommended routes are planned based on the object attribute information, and therefore, the candidate recommended routes can be sorted based on the priority corresponding to the object attribute information. For example, if the object attribute information priority is "fisherman" or "travel expert", then the recommended order of the candidate recommended routes is "most fishing spots along the way", "most fishing tackle shops along the way", and "most scenic spots along the way", and the routes with higher object attribute information priority will be recommended first.

[0246] S906: Sending the candidate recommended routes to the navigation terminal in the order of recommendation of the candidate recommended routes, so that the navigation terminal displays the candidate recommended routes in the order of recommendation of the candidate recommended routes.

[0247] In the embodiment of the present application, the candidate recommended routes are displayed on the navigation terminal according to the recommendation order obtained above, so that the target object can select them.

[0248] The embodiment of the present application recommends candidate routes through object attribute information and establishes corresponding route descriptions, and generates a recommended order of candidate recommended routes according to the priority corresponding to the object attribute information, so as to display the candidate recommended routes according to the recommended order, thereby achieving high personalization, deeply mining the user's navigation habits, generating travel routes that meet the user's preferences, and improving the navigation experience. And it also achieves comprehensive services: not only providing route planning, but also covering along-the-way and surrounding service recommendations to meet the diversified travel needs of users, so as to accurately grasp the user's navigation habits and real-time location information, and provide users with all-round and personalized travel service recommendations, significantly improving the practicality and user satisfaction of navigation software, and having broad application prospects and market value.

[0249] As an optional implementation, Fig. 9 is a flow chart of a method for updating a candidate recommended route provided in an embodiment of the present application, such as Fig. 9 As shown, after the above step S90, the method further includes:

[0250] S1: Acquire the current navigation route sent by the navigation terminal; the current navigation route is obtained by selecting candidate recommended routes on the navigation terminal by the target object; the current navigation route carries current navigation type information.

[0251] In an implementation manner of the present application, a current navigation route sent by the navigation terminal is obtained. The current navigation route is obtained by sending candidate recommended routes to the navigation terminal, the navigation terminal displaying the candidate recommended routes in a recommended order, and the target object selecting the candidate recommended routes; the current navigation route carries current navigation type information.

[0252] S3: When the current navigation type information and / or the real-time context information is updated, a new current position of the target object, new current navigation type information and new real-time context information are obtained.

[0253] In an embodiment of the present application, when any one of the current navigation type information and the real-time context information sends an update, new current navigation type information, new current navigation type information, and new real-time context information are obtained. Specifically, the current navigation type information represents the current navigation type, such as driving, cycling, walking, etc., and the current navigation type information update refers to the switching of navigation type, such as from driving to walking, etc. The way to identify the navigation type switch can be to identify the user switching from the car navigation to the mobile phone or other smart wearable devices, or from the mobile phone to the car navigation terminal or other navigation terminal update, or it can be the user's active switch, that is, actively requesting to change the navigation type. The real-time context information update refers to the schedule information and weather information, user mood, user current location, traffic conditions around the user's current location, etc. When any one of them is updated, it is regarded as a real-time context update.

[0254] As an optional implementation, after the above step S3, the above method further includes:

[0255] Determine whether the new real-time context information represents an update of travel intention; when the new real-time context information represents an update of travel intention, analyze the destination information based on the new real-time context information, historical travel data and object attribute information to determine the travel intention of the target object.

[0256] In the embodiment of the present application, when the new real-time context information represents the change of the user's destination information, or the change of daily schedule, or the change of weather that affects the travel intention, the destination information is analyzed again based on the new real-time context information, historical travel data and object attribute information to determine the travel intention of the target object. The destination information here may be the destination information carried in the initial navigation request, or it may be new destination information.

[0257] S5: Based on the new real-time situation information, the new current navigation type information and the travel intention, a destination recommendation process is performed on the new current location and the current navigation destination, a new candidate recommended destination is generated, and the new candidate recommended destination is sent to the navigation terminal so that the navigation terminal can display the new candidate recommended destination.

[0258] In an embodiment of the present application, a destination recommendation process is performed on the new current location and the current navigation destination based on the new real-time context information, the new current navigation type information and the travel intention, a new candidate recommended destination is obtained, and the new candidate recommended destination is sent to the navigation terminal so that the navigation terminal displays the new candidate recommended destination. Here, the method for performing destination recommendation is similar to S50 and will not be introduced here. The travel intention here is the travel intention corresponding to the navigation request when the new real-time context information does not represent an update of the travel intention; and is the new travel intention corresponding to the new real-time context information when the new real-time context information represents an update of the travel intention.

[0259] For example, when the user's travel intention is to go fishing, and the travel intention remains unchanged, when the user arrives at the destination, the user switches to walking navigation (if the navigation is started from the car computer, the data switches to the mobile phone when the user opens the mobile phone navigation). The walking navigation marks the distance of nearby fishing spots. Combined with the current weather conditions, it will switch to bridge holes on rainy days, shady fishing spots on high terrain and high temperatures, and sheltered fishing spots on windy days. (The fishing spot data source can be obtained by third-party software such as Diaoyubao).

[0260] For example, when the user's travel intention is to travel, but after the real-time situational information is updated, the new travel intention is to go to work, then the work address stored in the navigation software is recommended to the user; or based on the new travel intention and historical itinerary data, the object attribute information is used to recommend a destination, because the work address of the target object may not be unique.

[0261] S7: Acquire a new current navigation destination sent by the navigation terminal; the new current navigation destination is obtained by selecting a new candidate recommended destination for the target object on the navigation terminal.

[0262] S9: Based on the object attribute information and the object navigation preference model, perform route recommendation processing on the new current navigation destination, generate new candidate recommended routes, and send the new candidate recommended routes to the navigation terminal so that the navigation terminal displays the new candidate recommended routes.

[0263] In the embodiment of the present application, steps S7-S9 are similar to the recommended method of steps S70-S90 and will not be repeated here.

[0264] The embodiment of the present application re-recommends candidate destinations and candidate routes when the current navigation type information and / or real-time situation information is updated, thereby realizing real-time perception of user location and situation changes, and providing timely and applicable travel and service recommendations. Furthermore, by accurately grasping the user's navigation habits and real-time location information, it provides users with comprehensive and personalized travel service recommendations, significantly improving the practicality and user satisfaction of navigation software, and has broad application prospects and market value.

[0265] As an optional implementation, Fig.10 is a flow chart of a method for establishing an object navigation preference model provided in an embodiment of the present application, such as Fig.10 As shown, the method for establishing the object navigation preference model includes:

[0266] S01: Clean the historical travel data and convert the model data to obtain model data; the model data includes a training set and a test set;

[0267] In an embodiment of the present application, when establishing an object navigation preference model, the historical travel data is first cleaned and preprocessed to remove invalid data therein, and then the model data conversion process is performed to obtain the model data; the model data includes a training set and a test set. The purpose of the training set is to train the model, that is, to allow the model to learn the laws and patterns of the data through the training set. The characteristic is that the training set contains data samples that the model needs to learn, and these samples will be used to adjust the parameters of the model so that the model can make accurate predictions on new and unseen data. The quality of the training set directly affects the learning effect and performance of the model. The test set is used to evaluate the generalization ability of the model, that is, the performance of the model on unseen data. The characteristic is that it is a data sample independent of the training set. It is used to test the performance of the model after the model training is completed to ensure that the model does not just remember the data in the training set, but can be generalized to new data. The results of the test set are usually used as the final evaluation criteria for model performance.

[0268] S03: Perform feature selection on the training set to determine the model building data;

[0269] In the embodiment of the present application, after obtaining the training set, it is necessary to perform feature selection on the training set, and select the features that have a significant impact on the navigation decision behavior as the model building data, that is, select the features that can best distinguish the data as the model building data. The purpose of this process is to reduce the number of features, improve the efficiency of model training, reduce the risk of overfitting, and possibly improve the prediction performance of the model.

[0270] Optionally, algorithms such as information gain and Gini index can be used for feature selection. For example, when selecting information gain for feature selection, for the root node, the information gain of all candidate features is calculated. Information gain is based on the concept of information entropy, which is used to measure the reduction in the degree of chaos of a data set after using a certain feature to split the data. Specifically, it is necessary to calculate the entropy of the data set, calculate the information gain for each attribute, and select the feature with the highest information gain as the partitioning attribute of the current node. The root node refers to the starting point of the decision tree, and all data samples are first sent to the root node for processing. The decision tree uses entropy to evaluate the information gain of the feature to select the best split point.

[0271] The specific process may include: splitting the training set, calculating entropy, calculating information gain, etc. First, split the data set by selecting the feature with the highest information gain rate. Second, calculate the entropy of the data set. The higher the entropy, the higher the disorder of the data set; the lower the entropy, the closer the data set is to purity. Then, calculate the information gain of each attribute, and select the attribute with the largest information gain for splitting.

[0272] S05: Perform split feature processing and threshold division processing on the model building data to build a first preset object navigation preference model.

[0273] In the embodiment of the present application, before model construction, a decision tree algorithm is selected, such as CART (Classification and Regression Trees), ID3 (Iterative Dichotomy 3), C4.5 (an improved version of the ID3 algorithm), etc. Then split feature processing and partition threshold processing are performed. Split feature processing specifically refers to the process of determining how to use selected features to split data sets for each node based on the result of feature selection during the decision tree construction process. This includes determining the threshold of the split feature (for continuous variables) and how to distribute data to different child nodes according to the feature value. Partition threshold processing refers to the situation in the decision tree where, when the feature is continuous, one or more thresholds need to be determined to divide the range of feature values ​​into different intervals so that the data can be classified or grouped according to these intervals.

[0274] Taking C4.5 as an example, after obtaining the model building data or calculating the information gain, the construction process also includes:

[0275] 1. Calculate the information gain rate: C4.5 uses the information gain rate to select split features, which is calculated by dividing the information gain by the intrinsic information of the attribute (i.e. the entropy of the attribute value). This can reduce the preference for attributes with more values.

[0276] 2. Processing continuous attributes: C4.5 can process continuous attributes. It processes numerical data by calculating multiple split points of the attribute and finding the split point that can bring the maximum information gain rate.

[0277] 3. Handling missing values: C4.5 can handle the situation where there are missing values ​​in the data set. It handles this by estimating the contribution of the feature to the classification, rather than simply deleting the missing data.

[0278] S07: Pruning the first preset object navigation preference model to obtain a second preset object navigation preference model;

[0279] In the embodiment of the present application, when constructing the first preset object navigation preference model, in order to avoid the problem of overfitting, a post-pruning method is used to reduce the complexity of the tree, thereby improving the generalization ability. Pruning is achieved by deleting branches that contribute less to classification after constructing the tree. Specifically, it includes:

[0280] 1. Before pruning, classify the decision tree using the entire training set and calculate the classification error rate.

[0281] 2. Starting from the leaf nodes of the decision tree, recursively prune subtrees step by step upward until a certain stopping condition is reached or no further pruning is possible.

[0282] 3. After pruning, use the same training set to classify the simplified decision tree and calculate the classification error rate.

[0283] 4. Compare the error rates before and after pruning. If the error rate after pruning is higher, it means that the pruning operation has reduced the performance of the decision tree, so no pruning is performed; if the error rate after pruning is lower or the difference is not significant, pruning can be performed.

[0284] 5. Evaluation of loss function: Post-pruning is mainly achieved by minimizing the overall loss function of the decision tree. The loss function Lα(T) includes the empirical entropy Ht(T) and the regularization term α|T|, where α is the regularization parameter used to regulate the weight relationship between the empirical error term and the regularization term.

[0285] 6. Select the optimal subtree: The process of the post-pruning algorithm includes calculating the loss function L of the internal nodes from the leaf node to the top α (T t ), and update the regularization threshold α, and then get the α value set M of all nodes. Next, select the largest value α k , visit the internal nodes of the subtree from top to bottom, if Then pruning is performed and the predicted value of the leaf node is determined.

[0286] 7. Cross-validation to select the optimal subtree: Use cross-validation to select the optimal subtree T from the optional optimal subtree set. α .

[0287] S09: Based on the test set, adjust the second preset object navigation preference model to obtain an object navigation preference model.

[0288] Exemplarily, after obtaining the second preset object navigation preference model, the second preset object navigation preference model is evaluated and optimized based on the test set to obtain the final object navigation preference model. The decision tree converts each path from the root to the leaf into a rule, so that a rule list can be generated from the decision tree for predicting new data.

[0289] It should be noted that the above process introduces the object navigation preference model as a whole model. When necessary, the historical travel data can be classified according to the navigation type, so as to build a separate object navigation preference model based on each navigation type. The construction method is similar to the above process.

[0290] The embodiment of the present application processes historical travel data through a decision tree algorithm and constructs an object navigation preference model for recommending candidate routes, thereby deeply mining the user's navigation habits, generating travel routes that meet the user's preferences, and improving the navigation experience; and making different navigation recommendations based on the different travel routes selected by the user.

[0291] It should be noted that the above-mentioned object navigation preference model and object attribute information will be updated after each navigation to achieve more accurate prediction.

[0292] As an optional implementation method, the embodiment of the present application also provides another method for constructing an object navigation preference model, specifically, using a deep learning algorithm to analyze the user's facial expression to evaluate the user's driving status or emotional changes; or learning a large amount of navigation data to optimize the path planning and real-time scheduling functions of the navigation system, including the following steps:

[0293] 1. Data preprocessing: Preprocess the collected facial expression images or videos, including face detection, feature extraction, etc.; for other types of navigation data, corresponding preprocessing operations are also required, such as removing invalid data, etc., to finally obtain model training data.

[0294] 2. Build a deep learning model: Choose a suitable deep learning model architecture, such as convolutional neural network (CNN) for image recognition, recurrent neural network (RNN) for sequence data processing, etc.; and

[0295] Design the model's input and output layers, hidden layers, activation functions and other components.

[0296] 3. Model training: Use model training data to train the deep learning model, and adjust the weights and parameters of the model through forward propagation and back propagation algorithms to minimize the loss function.

[0297] Through the above steps, the object navigation preference model is finally obtained.

[0298] On the other hand, the embodiment of the present application further provides a navigation recommendation system, the navigation recommendation system includes navigation software and third-party software; the third-party software and the navigation software are data-interoperable; the navigation software is connected to a navigation server, and the navigation software includes a navigation habit analysis module. The navigation habit analysis module is used to construct object attribute information and an object navigation preference model.

[0299] In addition, the navigation software already has a constructed object navigation preference model and object attribute information, which are obtained based on the analysis of historical travel data and historical third-party software data by the navigation habit analysis module. Specifically, the historical travel data may include analysis of personalized features such as user travel patterns, preferred routes, and frequently visited places; the navigation software is also associated with third-party software, and the historical third-party software data includes weather, road conditions, users' historical dangerous driving data, historical violation data, historical browsing records, etc.

[0300] The navigation software is used to obtain historical travel data after receiving a navigation request from a user, and draw a navigation route based on the object navigation preference model and object attribute information, and mark high-risk driving areas, user historical violation points, scenic spots, parking lots, restaurants, etc. in the route with different colors. When the navigation software passes through the recommended destination or recommended area calculated by the object navigation preference model and object attribute information during the user's use, it will be highlighted, and the user can see the corresponding mark of the recommended destination or recommended area in the route overview.

[0301] On the other hand, Fig.11 As shown, the embodiment of the present application also provides a navigation recommendation device, the device comprising:

[0302] The first acquisition module 101 is used to obtain historical travel data of the target object, the current location of the target object and object attribute information in response to a navigation request of the target object; the navigation request carries destination information;

[0303] A determination module 103 is used to analyze the destination information based on the historical travel data and the object attribute information to determine the travel intention of the target object;

[0304] The first generating module 105 is used to perform destination recommendation processing on the destination information based on the travel intention, generate candidate recommended destinations, and send the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations;

[0305] The second acquisition module 107 is used to acquire the current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal;

[0306] The second generation module 109 is used to perform route recommendation processing on the current navigation destination and the current location based on the object attribute information and the object navigation preference model, generate candidate recommended routes, and send the candidate recommended routes to the navigation terminal so that the navigation terminal can display the candidate recommended routes; the object navigation preference model is obtained by processing the historical travel data.

[0307] As an optional implementation, the device includes a construction module of object attribute information, and the construction module of object attribute information includes:

[0308] A historical third-party software data acquisition unit is used to acquire historical third-party software data of a target object; the historical third-party software data represents historical data of third-party software associated with the navigation terminal;

[0309] A historical travel information determination unit, used for cleaning the historical travel data to remove invalid data in the historical travel data and obtain the historical travel information;

[0310] A historical interest information determination unit, used to classify and filter historical third-party software data to determine the historical interest information of the target object;

[0311] A data feature set construction unit, used to construct a data feature set based on historical travel information and historical interest information;

[0312] The object attribute information construction unit is used to perform standardization processing on the data feature set to obtain a standardized data feature set, and analyze the standardized data feature set based on cluster analysis to obtain object attribute information.

[0313] As an optional implementation, the object attribute information building module is further used to receive an object attribute information building request sent by the navigation terminal; the object attribute information building request carries preset object attribute information; the object attribute information building request is obtained by operating the target object on the navigation terminal;

[0314] It is also used to perform object attribute information construction processing on preset object attribute information to obtain object attribute information.

[0315] As an optional implementation, the determination module includes:

[0316] A first determination unit is used to filter and process the historical travel data according to the object attribute information to determine the historical travel data that matches the destination information;

[0317] A second determining unit is used to filter the historical interest information according to the object attribute information to determine the historical interest information that matches the destination information;

[0318] A real-time situation information acquisition unit, used to acquire real-time situation information of a target object; the real-time situation information includes schedule information and weather information;

[0319] The travel intention analysis unit is used to analyze the travel intention of the destination information based on the historical itinerary data matching the destination information, the historical interest information matching the destination information, and the real-time context information to obtain the travel intention.

[0320] As an optional implementation, the first generation module includes:

[0321] a first candidate recommended destination determining unit, configured to determine, based on the travel intention, a first candidate recommended destination within a preset distance of the destination information, and determine a second candidate recommended destination between the current location and the destination information;

[0322] a second candidate recommendation destination determining unit, configured to determine the first candidate recommendation destination and the second candidate recommendation destination as candidate recommendation destinations;

[0323] a candidate recommended destination sequence acquisition unit, configured to sort the candidate recommended destinations based on the distances between the candidate recommended destinations and the destination information to obtain a candidate recommended destination sequence;

[0324] The candidate recommended destination sequence sending unit is used to send the candidate recommended destination sequence to the navigation terminal so that the navigation terminal can display the candidate recommended destination sequence.

[0325] As an optional implementation manner, the candidate recommended destination sequence sending unit includes:

[0326] A recommendation reason determination unit, used to determine the recommendation reason corresponding to each candidate recommended destination based on the object attribute information and the travel intention;

[0327] The recommendation reason sending unit is used to send the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations to the navigation terminal, so that the navigation terminal can display the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations.

[0328] As an optional implementation, the second generation module includes:

[0329] A route description generating unit, for performing route recommendation processing on a current navigation destination and a current location based on an object navigation preference model, real-time context information and object attribute information, and generating preset candidate recommended routes and route descriptions corresponding to the preset candidate recommended routes;

[0330] A recommended location determination unit along the way, used to determine the recommended locations along the way corresponding to each preset candidate recommended route based on the travel intention and the current navigation destination;

[0331] A candidate recommended route construction unit, configured to construct a candidate recommended route based on preset candidate recommended routes, recommended places along the route corresponding to each preset candidate recommended route, and route descriptions corresponding to each preset candidate recommended route;

[0332] A priority determination unit, used to determine the priority corresponding to the object data information;

[0333] A recommended order generating unit for candidate recommended routes, used to sort the candidate recommended routes according to the priorities corresponding to the object attribute information, and obtain a recommended order for the candidate recommended routes;

[0334] The candidate recommended route sending unit is further used to send the candidate recommended routes to the navigation terminal in the recommended order of the candidate recommended routes, so that the navigation terminal displays the candidate recommended routes in the recommended order of the candidate recommended routes.

[0335] As an optional implementation, the priority determination unit is further used to prioritize the object attribute information based on historical travel data and historical third-party software data to obtain the priority corresponding to the object attribute information.

[0336] As an optional implementation manner, the priority determination unit is further configured to receive a priority sorting request for the object attribute information sent by the navigation terminal, perform priority sorting on the object attribute information, and obtain the priority corresponding to the object attribute information.

[0337] As an optional implementation, the device further includes a recommendation update module, which includes:

[0338] A first acquisition unit is used to acquire a current navigation route sent by the navigation terminal; the current navigation route is obtained by the target object selecting a candidate recommended route on the navigation terminal; the current navigation route carries current navigation type information;

[0339] A second acquisition unit, configured to acquire a new current position of the target object, new current navigation type information and new real-time context information when the current navigation type information and / or the real-time context information is updated;

[0340] a destination recommendation unit, configured to perform destination recommendation processing for a new current location and a current navigation destination based on new real-time scenario information, new current navigation type information and travel intention, generate a new candidate recommended destination, and send the new candidate recommended destination to the navigation terminal so that the navigation terminal displays the new candidate recommended destination;

[0341] A third acquisition unit is used to acquire a new current navigation destination sent by the navigation terminal; the new current navigation destination is obtained by selecting a new candidate recommended destination on the navigation terminal by the target object;

[0342] The route recommendation unit is used to perform route recommendation processing on the new current navigation destination based on object attribute information and the object navigation preference model, generate new candidate recommended routes, and send the new candidate recommended routes to the navigation terminal so that the navigation terminal can display the new candidate recommended routes.

[0343] As an optional implementation, the device further includes an object navigation preference model construction module, and the object navigation preference model construction module includes:

[0344] A model data acquisition unit is used to perform cleaning processing and model data conversion processing on historical travel data to obtain model data; the model data includes a training set and a test set;

[0345] A model building data determination unit, used to perform feature selection processing on the training set to determine the model building data;

[0346] A first preset model building unit, used for performing split feature processing and division threshold processing on the model building data to build a first preset object navigation preference model;

[0347] A second preset model building unit, configured to perform pruning processing on the first preset object navigation preference model to obtain a second preset object navigation preference model;

[0348] The object navigation preference model construction unit is used to adjust the second preset object navigation preference model based on the test set to obtain the object navigation preference model.

[0349] It should be noted that the navigation recommendation device embodiment provided in the present application and the above-mentioned navigation recommendation method embodiment are based on the same inventive concept.

[0350] An embodiment of the present application also provides an electronic device for navigation recommendation, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement a navigation recommendation method provided in any of the above embodiments.

[0351] An embodiment of the present application also provides a computer-readable storage medium, which can be set in a terminal to store at least one instruction or at least one program for implementing a navigation recommendation method in a method embodiment, and the at least one instruction or at least one program is loaded and executed by a processor to implement the navigation recommendation method provided in the above method embodiment.

[0352] Optionally, in the embodiment of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the above storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0353] The memory of the embodiment of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, applications required for functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0354] The present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device performs the above method embodiment to provide a navigation recommendation method.

[0355] The method embodiments provided in the embodiments of the present application can be executed in a terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Fig.12 is a hardware structure block diagram of a server of a navigation recommendation method provided in an embodiment of the present application. Fig.12As shown, the server 400 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 410 (the central processing unit 410 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 430 for storing data, and one or more storage media 420 (such as one or more mass storage devices) for storing application programs 423 or data 422. Among them, the memory 430 and the storage medium 420 can be short-term storage or permanent storage. The program stored in the storage medium 420 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 410 can be configured to communicate with the storage medium 420 and execute a series of instruction operations in the storage medium 420 on the server 400. The server 400 may also include one or more power supplies 460, one or more wired or wireless network interfaces 450, one or more input and output interfaces 440, and / or one or more operating systems 421, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0356] The input / output interface 440 may be used to receive or send data via a network. The specific example of the network may include a wireless network provided by a communication provider of the server 400. In one example, the input / output interface 440 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 440 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0357] It can be understood by those skilled in the art that Fig.12 The structure shown is only for illustration and does not limit the structure of the above electronic device. Fig.12 More or fewer components as shown, or with Fig.12 Different configurations are shown.

[0358] It should be noted that the above-mentioned sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0359] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and server embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0360] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0361] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A navigation recommendation method, characterized in that: The method comprises: In response to a navigation request of a target object, acquiring historical travel data of the target object, a current location of the target object, and object attribute information; the navigation request carries destination information; Analyzing the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object; Based on the travel intention, performing destination recommendation processing on the destination information to generate candidate recommended destinations, and sending the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations; Acquire a current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal; Based on the object attribute information and the object navigation preference model, route recommendation processing is performed on the current navigation destination and the current location, candidate recommended routes are generated, and the candidate recommended routes are sent to the navigation terminal so that the navigation terminal displays the candidate recommended routes; the object navigation preference model is obtained by processing the historical travel data.

2. The method according to claim 1, characterized in that The method for constructing the object attribute information includes: Acquire historical third-party software data of the target object; the historical third-party software data represents historical data of third-party software associated with the navigation terminal; Cleaning the historical travel data to remove invalid data in the historical travel data to obtain historical travel information; Classifying and filtering the historical third-party software data to determine the historical interest information of the target object; Constructing a data feature set according to the historical travel information and the historical interest information; The data feature set is standardized to obtain a standardized data feature set, and the standardized data feature set is analyzed based on cluster analysis to obtain the object attribute information.

3. The method according to claim 1, characterized in that The method for constructing the object attribute information further includes: receiving an object attribute information establishment request sent by the navigation terminal; the object attribute information establishment request carries preset object attribute information; the object attribute information establishment request is obtained by operating the target object on the navigation terminal; The preset object attribute information is subjected to object attribute information construction processing to obtain the object attribute information.

4. The method according to claim 2, characterized in that: The step of analyzing the destination information according to the historical travel data and the object attribute information to determine the travel intention of the target object includes: Filtering the historical travel data according to the object attribute information to determine the historical travel data that matches the destination information; filtering the historical interest information according to the object attribute information to determine the historical interest information that matches the destination information; Acquire real-time situation information of the target object; the real-time situation information includes schedule information and weather information; Based on the historical travel data matching the destination information, the historical interest information matching the destination information, and the real-time context information, the purpose information is analyzed for travel intention to obtain the travel intention.

5. The method according to claim 1, characterized in that The performing destination recommendation processing on the destination information based on the travel intention, generating candidate recommended destinations, and sending the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations, includes: Based on the travel intention, determining a first candidate recommended destination within a preset distance of the destination information, and determining a second candidate recommended destination between the current location and the destination information; determining the first candidate recommended destination and the second candidate recommended destination as the candidate recommended destinations; sorting the candidate recommended destinations based on the distances between the candidate recommended destinations and the destination information to obtain a candidate recommended destination sequence; The candidate recommended destination sequence is sent to the navigation terminal, so that the navigation terminal displays the candidate recommended destination sequence.

6. The method according to claim 5, characterized in that The sending the candidate recommended destination sequence to the navigation terminal so that the navigation terminal displays the candidate recommended destination sequence includes: Determining, based on the object attribute information and the travel intention, a recommendation reason corresponding to each of the candidate recommended destinations; The candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations are sent to the navigation terminal, so that the navigation terminal displays the candidate recommended destinations and the recommendation reasons corresponding to the candidate recommended destinations.

7. The method according to claim 4, characterized in that The method of performing route recommendation processing on the current navigation destination and the current location based on the object attribute information and the object navigation preference model, generating candidate recommended routes, and sending the candidate recommended routes to the navigation terminal so that the navigation terminal displays the candidate recommended routes includes: Based on the object navigation preference model, the real-time scenario information and the object attribute information, performing route recommendation processing on the current navigation destination and the current location, generating a preset candidate recommended route and a route description corresponding to each of the preset candidate recommended routes; Based on the travel intention and the current navigation destination, determining recommended locations along the way corresponding to each of the preset candidate recommended routes; constructing the candidate recommended routes based on the preset candidate recommended routes, the recommended places along the way corresponding to the preset candidate recommended routes, and the route descriptions corresponding to the preset candidate recommended routes; Determining the priority corresponding to the object data information; Sorting the candidate recommended routes according to the priorities corresponding to the object attribute information to obtain a recommended order of the candidate recommended routes; The candidate recommended routes are sent to the navigation terminal according to the recommended order of the candidate recommended routes, so that the navigation terminal displays the candidate recommended routes according to the recommended order of the candidate recommended routes.

8. The method according to claim 7, characterized in that The determining the priority corresponding to the object data information includes: Based on the historical travel data and the historical third-party software data, the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

9. The method according to claim 7, characterized in that: The determining of the priority corresponding to the object data information further includes: A priority sorting request for the object attribute information sent by the navigation terminal is received, and the object attribute information is prioritized to obtain the priority corresponding to the object attribute information.

10. The method according to claim 7, characterized in that After sending the candidate recommended routes to the navigation terminal in the recommended order of the candidate recommended routes so that the navigation terminal displays the candidate recommended routes in the recommended order of the candidate recommended routes, the method further includes: Acquire a current navigation route sent by the navigation terminal; the current navigation route is obtained by the target object selecting the candidate recommended route on the navigation terminal; the current navigation route carries current navigation type information; When the current navigation type information and / or the real-time context information is updated, obtaining a new current position of the target object, new current navigation type information and new real-time context information; Based on the new real-time situation information, the new current navigation type information and the travel intention, perform destination recommendation processing on the new current location and the current navigation destination, generate a new candidate recommended destination, and send the new candidate recommended destination to the navigation terminal so that the navigation terminal displays the new candidate recommended destination; Acquire a new current navigation destination sent by the navigation terminal; the new current navigation destination is obtained by selecting the new candidate recommended destination by the target object on the navigation terminal; Based on the object attribute information and the object navigation preference model, a route recommendation process is performed on the new current navigation destination to generate a new candidate recommended route, and the new candidate recommended route is sent to the navigation terminal so that the navigation terminal displays the new candidate recommended route.

11. The method according to claim 1, characterized in that: The method for establishing the object navigation preference model includes: The historical travel data is cleaned and converted into model data to obtain model data; the model data includes a training set and a test set; Performing feature selection processing on the training set to determine model building data; Performing split feature processing and division threshold processing on the model building data to build a first preset object navigation preference model; Pruning the first preset object navigation preference model to obtain a second preset object navigation preference model; Based on the test set, the second preset object navigation preference model is adjusted to obtain the object navigation preference model.

12. A navigation recommendation device, characterized in that: The device comprises: A first acquisition module is used to obtain historical travel data of the target object, the current location of the target object and object attribute information in response to a navigation request of the target object; the navigation request carries destination information; A determination module, configured to analyze the destination information according to the historical travel data and the object attribute information, and determine the travel intention of the target object; A first generating module is used to perform destination recommendation processing on the destination information based on the travel intention, generate candidate recommended destinations, and send the candidate recommended destinations to the navigation terminal so that the navigation terminal displays the candidate recommended destinations; A second acquisition module is used to acquire a current navigation destination sent by the navigation terminal; the current navigation destination is obtained by the target object selecting the candidate recommended destination on the navigation terminal; The second generation module is used to perform route recommendation processing on the current navigation destination and the current location based on the object attribute information and the object navigation preference model, generate candidate recommended routes, and send the candidate recommended routes to the navigation terminal so that the navigation terminal displays the candidate recommended routes; the object navigation preference model is obtained by processing the historical travel data.

Images