Route recommendation method and device, medium and product

Abstract

The embodiment of the invention provides a route recommendation method and device, a medium and a product. The method comprises the steps that recommended routes are determined according to navigation request data of a user and route preference labels, the route preference labels are determined according to the difference degree between a target recommended route and other recommended routes in the recommended routes, and the target recommended route and the other recommended routes are determined according to the deviation degree between the recommended routes and an actual driving route; the recommended route and the actual driving route are route data corresponding to historical navigation request data of the user; and sending the recommended route to the client, and displaying the recommended route in the client. The method is used for achieving the effects of improving route recommendation accuracy and meeting user preference requirements.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a route recommendation method, device, medium and product. Background Art

[0002] In the field of intelligent navigation and user behavior analysis, setting multiple preference themes and presenting specific interfaces to target users can effectively improve user experience and service accuracy. This process involves in-depth exploration and understanding of user preferences, with deep learning models playing a key role in identifying user preferences.

[0003] Current technology delegates the task of labeling to the user within the route preference interface design. Simultaneously, deep learning models are used to accurately identify user preferences based on historical navigation preference data, providing users with route planning services that better meet their personalized needs.

[0004] However, since users' behavior patterns are complex and changeable when traveling on a route, there is a problem that route recommendations are inaccurate and cannot meet user preference needs. Summary of the Invention

[0005] The embodiments of the present application provide a route recommendation method, device, medium, and product to improve the accuracy of route recommendations and meet user preference requirements.

[0006] In a first aspect, an embodiment of the present application provides a route recommendation method, comprising:

[0007] Get the user's navigation request data;

[0008] Determine a recommended route based on the user's navigation request data and route preference tags, where the route preference tags are determined based on the degree of difference between a target recommended route and other recommended routes in the recommended routes, and the target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual travel route, where the recommended route and the actual travel route are route data corresponding to the user's historical navigation request data;

[0009] The recommended route is sent to the client for display in the client.

[0010] In one possible implementation, the method further includes:

[0011] Determining the same routes between the recommended route and the actual traveled route, and the route lengths of the same routes;

[0012] Determine the degree of deviation between the recommended route and the actual route traveled based on the route length of the same route and the actual route length of the actual route traveled;

[0013] According to the degree of deviation between the recommended route and the actual driving route, a target recommended route and other recommended routes in the recommended route are determined.

[0014] In one possible implementation, the method further includes:

[0015] Determine the user's preference characteristics in different preference dimensions based on the difference between the target recommended route and other recommended routes in the recommended routes;

[0016] Determine the user's route preference label based on the user's preference characteristics in different preference dimensions.

[0017] In a possible implementation, determining the user's preference characteristics in different preference dimensions based on the degree of difference between the target recommended route and other recommended routes in the recommended routes includes:

[0018] According to the preset preference dimension, determining the characteristic extreme values of other route features in other recommended routes and the characteristic values of route features in the target recommended route features;

[0019] Determine a difference characteristic value according to characteristic extreme values of other route characteristics in other recommended routes and characteristic values of route characteristics in the target recommended route characteristics;

[0020] The user's preference characteristics under different preference dimensions are determined based on the difference characteristic values and the characteristic values of the route characteristics in the target recommended route characteristics.

[0021] In a possible implementation, determining the user's route preference label based on the user's preference characteristics in different preference dimensions includes:

[0022] Determine the difference between the actual driving route and the recommended route based on the user's preference characteristics in different preference dimensions;

[0023] Determine the user's route preference label based on the difference between the actual driving route and the recommended route;

[0024] or,

[0025] Cluster the user's preference features under different preference dimensions to obtain clustering results;

[0026] Based on the clustering results, the user's route preference label is determined.

[0027] In a possible implementation, when the navigation request data includes request time data and start and end point data, and the historical navigation request data includes historical request time data and historical start and end point data,

[0028] Determine recommended routes based on the user's navigation request data and route preference tags, including:

[0029] Determine a user preference model, which is trained based on the user's route preference tags and historical navigation request data;

[0030] The navigation request data is input into the user preference model to obtain the recommended route output by the user preference model.

[0031] In a second aspect, an embodiment of the present application provides a route recommendation method, applied to a client, the method comprising:

[0032] In response to the user's input operation, the navigation request data input by the user is sent to the server;

[0033] Receive and display the recommended route sent by the server, where the recommended route is determined by the server based on the navigation request data and the route preference tag. The route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route. The target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual travel route. The recommended route and the actual travel route are the route data corresponding to the user's historical navigation request data.

[0034] In a third aspect, an embodiment of the present application provides a route recommendation device, applied to a server, comprising:

[0035] The acquisition module is used to obtain the user's navigation request data;

[0036] a determination module, configured to determine a recommended route based on the user's navigation request data and a route preference tag, wherein the route preference tag is determined based on a degree of difference between a target recommended route and other recommended routes in the recommended route, wherein the target recommended route and other recommended routes are determined based on a degree of deviation between the recommended route and an actual travel route, wherein the recommended route and the actual travel route are route data corresponding to the user's historical navigation request data;

[0037] The sending module is used to send the recommended route to the client for display in the client.

[0038] In a fourth aspect, an embodiment of the present application provides a route recommendation device, applied to a client, and a method including:

[0039] The response module is used to send the navigation request data input by the user to the server in response to the user's input operation;

[0040] The display module is used to receive and display the recommended route sent by the server, wherein the recommended route is determined by the server based on the navigation request data and the route preference tag, the route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route, the target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are the route data corresponding to the user's historical navigation request data.

[0041] In a fifth aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0042] Memory stores computer-executable instructions;

[0043] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0044] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementation methods of the first aspect.

[0045] In a seventh aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0046] The route recommendation method, device, medium and product provided in the embodiments of the present application obtain the user's navigation request data; determine the recommended route based on the user's navigation request data and route preference tag, the route preference tag is determined according to the degree of difference between the target recommended route and other recommended routes in the recommended route, the target recommended route and other recommended routes are determined according to the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are route data corresponding to the user's historical navigation request data; send the recommended route to the client, and display it in the client, so that when recommending the recommended route, the target recommended route in the historical navigation request data that is close to the actual driving route can be determined, and then the difference between the target recommended route and the other recommended routes is obtained to determine the route preference tag that represents the user's behavior pattern, and finally, based on the route preference tag and the user's navigation request data, determine a recommended route that better meets the user's personalized needs, thereby achieving the effect of improving the accuracy of route recommendations and meeting user preference needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0048] Figure 1 Schematic diagram of the model training method and route recommendation method provided in this application;

[0049] Figure 2 A flowchart of the model training method provided in this application;

[0050] Figure 3 A flowchart of the route recommendation method provided for this application;

[0051] Figure 4 A schematic diagram of the structure of the model training device provided in this application;

[0052] Figure 5 A schematic diagram of the structure of the route recommendation device provided in this application;

[0053] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application.

[0054] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0055] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0056] First, let’s explain the terms involved in this application:

[0057] Route recommendation is a process that uses multiple data such as maps, real-time traffic, and user history in traffic and travel scenarios, and uses algorithms such as Dijkstra to calculate in complex road networks. It provides users with the best travel route suggestions from the starting point to the end point based on their location, destination, and personalized needs such as fastest, most economical, and most comfortable. It is widely used in navigation and taxi platforms to help users travel efficiently.

[0058] In the existing technology, there are mainly two ways to determine user route preferences. The first is manual setting by the user. In common navigation apps, users can manually select navigation route preferences in the settings options, such as setting them to "avoid congestion priority", "highway priority", "shortest distance priority", etc. This method relies on users to actively clarify their own needs, but it has obvious disadvantages. On the one hand, users need to spend time and energy to manually operate the settings, which is a cumbersome process. If the user does not have a clear understanding of their own preferences, it is difficult to set options that meet actual needs; on the other hand, when the user's travel scenario changes, such as travel time, destination, and traffic conditions, manual adjustments need to be made, which lacks flexibility and intelligence.

[0059] The second approach is to leverage modeling. Some technologies present a route preference interface to the target user by setting multiple preference themes, offloading the labeling task to the user. Deep learning models are then used to identify user preferences based on their historical navigation preferences. However, this approach lacks temporal and spatial attributes, making it less adaptable to complex and changing user behavior patterns. Users may have different route preferences at different times and locations, which can affect the accuracy and adaptability of navigation route recommendations.

[0060] The route recommendation method provided in this application analyzes the user's historical navigation data and compares the differences between the target recommended route (the recommended route closest to the actual driving path) and other unselected recommended routes to accurately identify the user's implicit preferences. This effectively eliminates the interference of random selection and focuses on the key features that truly influence the user's decision-making, thereby generating accurate route preference labels. When a user initiates a new request, the real-time traffic conditions are combined with these deeply mined preference features to achieve personalized route recommendations, thereby improving the accuracy of route recommendations and meeting user preference needs.

[0061] Figure 1 A schematic diagram of the route recommendation method provided in this application, such as Figure 1 As shown, the specific application scenarios of this application can be client and server, wherein the client and server work together to help users achieve accurate navigation. The client is equipped with a navigation system. When the user specifies the starting and ending points and initiates a navigation request, the navigation system in the client will quickly capture the starting and ending point information and time information, and send these key data to the server. The server can configure a user preference model, which can accurately output the user's preferred recommended route based on the received starting and ending point information, time information and route preference tags through complex algorithm calculations. Subsequently, the server will transmit the generated preferred recommended route back to the client, and the client will then clearly display the preferred recommended route in the interface, making it convenient for users to intuitively view and plan their itinerary.

[0062] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0063] Figure 2 A flowchart of the route recommendation method provided for this application is shown in FIG. Figure 2 As shown, the method includes:

[0064] S201: Obtain user's navigation request data.

[0065] Here, a user can refer to an individual user or a specific group of people. For example, when the user is an individual user, the object can be an office worker who frequently uses navigation software for travel. When the user is a specific group of people, the object can be a tour group that frequently travels together in a certain area.

[0066] Navigation request data may refer to a set of route planning related information entered by a user when using a navigation service, which may include request time data and start and end point data, where:

[0067] Request time data refers to the time when a user initiates a navigation request. This time can be accurate to the year, month, day, hour, minute, and second. For example, a request is initiated at d:00 am, e:00 am, and f:00 am on year a, month b, and c:00 am. This data can reflect the user's daily travel patterns, whether it's a regular weekday commute or a leisure trip on the weekend. Furthermore, this data can reflect special dates, such as holidays and anniversaries, where user travel needs may differ from those on weekdays.

[0068] The starting and ending point data may refer to the starting and ending points, and the location type. The starting and ending points may refer to the starting location and target location involved in the user's activity. The location type may refer to the classification of the starting and ending points according to different functions and properties. For example, the location type may include residential locations, work locations, commercial locations, transportation locations, educational locations, medical locations, leisure and entertainment locations, etc.

[0069] S202. Determine a recommended route based on the user's navigation request data and route preference tag. The route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route. The target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route. The recommended route and the actual driving route are route data corresponding to the user's historical navigation request data.

[0070] Recommended routes may refer to multiple suggested driving routes generated by the navigation system through algorithmic calculations based on the user's navigation request data and real-time traffic information. The actual driving route may refer to the route the user actually drove when using the navigation service. The target recommended route within a recommended route may refer to the route closest to the actual driving route, and other recommended routes may refer to recommended routes other than the target recommended route.

[0071] When using navigation services, a user's recommended route, actual travel route, target recommended route, and other recommended routes can be determined based on the user's historical navigation request data. This data refers to a complete record of the user's past navigation service use. This record can include key information such as the starting and ending locations of each navigation, travel mode selection, and request time. It can also include actual user behavior data such as GPS trajectory data, route deviations, and travel time, as well as user adjustments and feedback on recommended routes.

[0072] The target recommended route and other recommended routes in the recommended routes can be determined based on the degree of deviation between the recommended route and the actual travel route. The degree of deviation between the recommended route and the actual travel route can represent the inconsistency rate between the recommended route and the actual travel route, that is, it can represent the degree of path overlap between the recommended route and the actual travel route. In this embodiment of the present application, the recommended route with the highest degree of path overlap can be the target recommended route, and the other recommended routes can be other recommended routes.

[0073] In this embodiment of the present application, the method for determining the target recommended route and other recommended routes may include:

[0074] Determining the same routes between the recommended route and the actual traveled route, and the route lengths of the same routes;

[0075] Determine the degree of deviation between the recommended route and the actual route traveled based on the route length of the same route and the actual route length of the actual route traveled;

[0076] According to the degree of deviation between the recommended route and the actual driving route, a target recommended route and other recommended routes in the recommended route are determined.

[0077] The term "same route" may refer to the overlapping portion of the recommended route and the actual route, and the length of the same route may refer to the length of the overlapping portion. In some embodiments, the latitude and longitude sequences of the user's actual trajectory and the recommended route may be matched to a road network, and the same route and length of the same route may be determined based on the same road segment ID.

[0078] The ratio of the length of the route that overlaps with the actual length of the route can be used to determine the degree of deviation between the recommended route and the actual route. That is, the degree of deviation between the recommended route and the actual route can meet the following requirements:

[0079] R=1-D 相同 / D 实际 ;

[0080] Among them, R represents the degree of deviation between the recommended route and the actual driving route, which can be expressed as the inconsistency rate between the recommended route and the actual driving route, D 相同 It can represent the route length of the same route in the recommended route and the actual driving route, D 实际 The actual route length of the actual driving route can be represented.

[0081] Therefore, the inconsistency rate between each recommended route and the actual driving route can be determined by the size of R, and the recommended route with the smallest inconsistency rate is regarded as the target recommended route that meets the deviation degree requirement, and the other recommended routes are regarded as other recommended routes that do not meet the deviation degree requirement.

[0082] After determining the target recommended route and the other recommended routes among the recommended routes, a route preference label may be determined according to the degree of difference between the target recommended route and the other recommended routes among the recommended routes.

[0083] Among them, route preference labels can refer to markers used to identify users' unique preferences in route selection. For example, when a route preference label is characterized as "time priority," it means that in past navigation, users often choose routes with the shortest estimated travel time, even if that route may be slightly longer or have higher tolls. When a route preference label is characterized as "distance priority," it indicates that users tend to choose routes with the shortest distance and are less concerned about travel time or road conditions. When a route preference label is characterized as "economical," it indicates that users often prefer routes that are toll-free and do not take highways, and are willing to accept longer travel times or less-than-ideal road conditions to save money. When a route preference label is characterized as "comfortable road conditions," it indicates that users often choose routes with good road conditions and less congestion, even if the route distance or time is not optimal. Route preference labels can intuitively reflect users' route selection preferences in different dimensions such as time, distance, cost, and road conditions.

[0084] The degree of difference between the target recommended route and other recommended routes can characterize the degree of proximity between the target recommended route and other recommended routes in multiple dimensions. For example, in terms of distance, the difference in length of each route is calculated. The larger the difference, the higher the degree of distance difference. In terms of direction, the turning angles of key nodes are compared. If the angle difference is large, it means that the direction difference is obvious. In terms of passing locations, the lower the overlap of the route passing locations and the more different locations, the greater the degree of difference. In terms of driving time, the greater the difference between the time taken by each route and the actual driving time, the higher the degree of time difference. The difference in road conditions is reflected in the different types of roads (such as main roads, side roads) and congestion conditions (congested, unobstructed). The greater the difference in road conditions between routes, the higher the degree of road condition difference. In this way, the degree of difference between different routes can be reflected.

[0085] In some embodiments, the degree of closeness between the target recommended route and other recommended routes in different dimensions can be represented by difference features. For example, the degree of difference in time between the target recommended route and other recommended routes can be represented as the difference in their estimated travel times.

[0086] Based on this, by determining the proximity between the target recommended route and other recommended routes in different dimensions, the user's choices of different preferences when driving can be obtained, and the user's route preference label can be determined.

[0087] In an embodiment of the present application, a method for determining a route preference label based on the degree of difference between a target recommended route and other recommended routes in a recommended route may include:

[0088] Determine the user's preference characteristics in different preference dimensions based on the difference between the target recommended route and other recommended routes in the recommended routes;

[0089] Determine the user's route preference label based on the user's preference characteristics in different preference dimensions.

[0090] Among them, the user's preference characteristics in different preference dimensions can refer to the user's personalized route selection tendency characteristics (such as time sensitivity coefficient, high-speed avoidance index, etc.) quantified and analyzed by comparing the differences between the target recommended route and other recommended routes in dimensions such as path efficiency, road type, economic cost, and driving experience.

[0091] In some embodiments, the degree of difference between the target recommended route and other recommended routes in the recommended routes may include the degree of difference in different preference dimensions, such as the distance dimension, the time dimension, the traffic light dimension, etc. After determining the degree of difference in different preference dimensions, the user's preference characteristics in different preference dimensions can be determined.

[0092] For example, compared with other recommended routes, the target recommended route has a shorter estimated travel time, and the time difference with other recommended routes is as large as 10 minutes. This indicates that the user may be more inclined to choose a shorter route in terms of time dimension, and has a time-priority preference characteristic. The distance of the target recommended route is in the middle value, and the distance difference with the other two recommended routes is 2 kilometers. The difference is not significant, and it is not clear for the time being that the user has a clear preference in the distance dimension. The target recommended route has good road conditions, which is quite different from other recommended routes, indicating that the user may prefer routes with better road conditions. The target recommended route does not take the highway and is free of charge, which is in sharp contrast to the recommended routes that have highway sections and are toll-free, reflecting that the user may have a preference to avoid highway sections and tolls.

[0093] After determining the user's preference characteristics across different preference dimensions, the user's route preference label is determined by analyzing these characteristics across the different preference dimensions. For example, as shown in the above example, when selecting a route, the user's route preference labels can be categorized as "time priority, comfortable road conditions, and economical." In some embodiments, to facilitate processing of route preference labels, the route preference labels can be displayed as normalized vectors to determine the user's route preference label.

[0094] When determining the preference characteristics, the determination can be made based on the characteristic values in the recommended routes. In the embodiment of the present application, the user's preference characteristics in different preference dimensions are determined based on the degree of difference between the target recommended route and other recommended routes in the recommended routes, including:

[0095] According to the preset preference dimension, determining the characteristic extreme values of other route features in other recommended routes and the characteristic values of route features in the target recommended route features;

[0096] Determine a difference characteristic value according to characteristic extreme values of other route characteristics in other recommended routes and characteristic values of route characteristics in the target recommended route characteristics;

[0097] The user's preference characteristics under different preference dimensions are determined based on the difference characteristic values and the characteristic values of the route characteristics in the target recommended route characteristics.

[0098] The preset preference dimension may refer to a preset type of route feature. Based on the preset preference dimension, corresponding route features and their characteristic values can be extracted from the route data. For example, features such as distance, travel time, number of traffic lights, and road conditions can be extracted.

[0099] The other route features in other recommended routes may refer to route features related to other recommended routes, and the route features in target recommended route features may refer to route features related to the target recommended route features.

[0100] The characteristic extreme values of other route features in other recommended routes may refer to the characteristic maximum values and characteristic minimum values corresponding to other route features in all other recommended routes. For example, when there are 3 other recommended routes and the other route features are the number of traffic lights, which are 5, 6, and 7 respectively, then the characteristic extreme values of other route features in other recommended routes are 5 and 7.

[0101] The difference characteristic value may be the difference between the characteristic extreme values of other route characteristics in other recommended routes and the characteristic value of the route characteristics in the target recommended route characteristics. For example, if the number of traffic lights representing the route characteristics in the target recommended route is 5, the difference characteristic value may be 0 or -2. In this embodiment of the present application, the difference characteristic value includes the characteristic value of the route characteristics in the target recommended route characteristics, the maximum characteristic difference value, and the minimum characteristic difference value between the characteristic value of the route characteristics and the characteristic extreme values of other route characteristics.

[0102] For example, there are five recommended routes with travel times of 25 minutes, 27 minutes, 30 minutes, 28 minutes, and 50 minutes, respectively. Recommended route 1, with a travel time of 25 minutes, has the lowest discrepancy between the user's actual trajectory and the navigation. The maximum and minimum travel times for this group of candidate routes are 50 minutes, and 25 minutes, respectively. The difference between the maximum and minimum travel times for recommended route 1 and recommended route 1 is -25 minutes and 0 minutes, respectively, indicating that recommended route 1, the route the user took, has the shortest travel time among the candidate routes. A similar approach is used for other features. If the route feature has 50 dimensions, calculating the difference between each feature and the maximum and minimum values within the group yields a 150-dimensional route preference feature.

[0103] In this embodiment of the present application, determining the user's route preference label based on the user's preference characteristics in different preference dimensions includes:

[0104] Determine the difference between the actual driving route and the recommended route based on the user's preference characteristics in different preference dimensions;

[0105] Determine the user's route preference label based on the difference between the actual driving route and the recommended route;

[0106] or,

[0107] Cluster the user's preference features under different preference dimensions to obtain clustering results;

[0108] Based on the clustering results, the user's route preference label is determined.

[0109] When determining a user's route preference tag, the actual route traveled and the recommended route can be compared based on the user's preference characteristics across different dimensions, such as time, distance, road conditions, and highway tolls. For example, if the user's preference characteristics indicate a preference for shorter routes, the comparison will focus on the difference in travel time between the actual route and the recommended route; if the user prefers routes with better road conditions, the road conditions of the two routes will be compared. Through this multi-dimensional comparative analysis, the differences between the actual route traveled and the recommended route in various aspects can be determined. Based on these specific differences, the user's overall tendency in route selection can be determined, and a route preference tag that accurately summarizes their route selection habits can be determined.

[0110] Alternatively, the user's preference characteristics across different preference dimensions, such as time, distance, road conditions, and tolls, can be aggregated and processed using clustering methods such as K-Medoids clustering and hierarchical clustering to obtain the user's route preference label. Taking K-Medoids clustering as an example, k representative objects are randomly selected as the initial cluster centers, and the remaining objects are assigned to the cluster closest to the representative objects. During each iteration, a non-central point is randomly selected to replace one of the original central points, and the clustering results are recalculated. If the effect improves, the replacement is retained; otherwise, the original central point is restored. The iteration ends when the replacement fails to improve the effect. Through clustering, route preference characteristics can be divided into k clusters with similar feature preferences within a cluster and significant differences between clusters. Based on the characteristic features of each cluster, the corresponding preference label can be obtained.

[0111] In an embodiment of the present application, preference features can be obtained based on the difference between the user route features and the maximum / minimum values of the features within the group, and then the route preference label can be determined by formulating rules or clustering methods.

[0112] When formulating rules, it's important to consider the differences between the user's target route and other recommended routes within the group, taking into account multiple dimensions such as time, distance, highways, tolls, and congestion. For example, if a user's chosen route has the fastest eta at a given moment and no other significant advantages, then they prefer faster travel time. If a user chooses a toll-free route that doesn't take highways and doesn't have the fastest eta, then they prefer to avoid tolls and are willing to accept a certain time difference.

[0113] After determining the user's route preference tag, a recommended route can be selected from the initial recommended routes generated by the user's navigation request data based on the route preference tag. That is, an initial set of candidate routes can be generated based on the navigation request (such as starting point, end point, and travel mode), and then the established route preference tag can be used as a filtering condition for optimization.

[0114] Recommended routes can also be generated based on a user preference model. This involves taking the route preference tag and navigation request data as input features. The user preference model dynamically analyzes the underlying features of the route preference tag and navigation request data and their relevance to the current scenario, directly generating a recommended route that incorporates personalized preferences.

[0115] When the navigation request data includes request time data and start and end point data, and the historical navigation request data includes historical request time data and historical start and end point data, in an embodiment of the present application, determining a recommended route based on the user's navigation request data and route preference tags includes:

[0116] Determine a user preference model, which is trained based on the user's route preference tags and historical navigation request data;

[0117] The navigation request data is input into the user preference model to obtain the recommended route output by the user preference model.

[0118] The request time data includes time granularity information and date attribute information. The time granularity information may include information such as day of the week, date, hour, minute, etc., and the location type information may include information such as whether it is a working day or a holiday.

[0119] The starting and ending point data include the location information of the starting and ending points and the location type information. The location information of the starting and ending points may include the latitude and longitude of the starting and ending points, geohash grid and other information, and the location type information may include the location type and other information.

[0120] The user preference model can be a machine learning model, including but not limited to XGBoost, neural network models, and other models. For example, XGBoost is an open-source gradient boosting algorithm that builds multiple decision trees based on the time and location characteristics of a user's historical navigation trips. It increases the weight of incorrectly predicted samples and gradually corrects the previous erroneous samples, thereby improving prediction performance.

[0121] In the embodiments of the present application, user preferences may vary under different temporal and spatial conditions. For example, during the weekday, users may prefer faster routes when commuting, but on weekends, they are less sensitive to travel time, are willing to accept a certain time difference, and prefer routes that are comfortable and safe. Spatially, user preferences also differ significantly when commuting to work or to a place of entertainment. Therefore, when training a user preference model, navigation request data, including request time data and start and end point data, can be input into the user preference model to obtain the output of the user preference model. Based on the output and the route preference labels, a loss function for the user preference model is determined, and the user preference model is adjusted until the loss function converges.

[0122] Therefore, the user preference model can explore the path selection patterns of users in different scenarios. When a new navigation request is input, the model will combine real-time traffic conditions, user tag weights (such as a time preference coefficient of 0.8) and contextual features (such as whether it is currently commuting time), and directly output personalized recommended routes through algorithms such as deep neural networks or ensemble learning. The resulting personalized recommended routes can inherit the user's historical preferences (such as sticking to familiar routes) and dynamically adapt to new needs (such as temporarily adjusting waypoints), achieving more accurate route matching than rule-based screening.

[0123] S203: Send the recommended route to the client for display in the client.

[0124] The route recommendation method provided in the embodiment of the present application uses actual driving routes and recommended route data to mine the target recommended route with the lowest inconsistency rate, as well as the route characteristics of the target recommended route, such as distance, travel time, etc., and at the same time calculates the difference characteristics between it and the extreme values of other route characteristics. In this way, the route preference characteristics are obtained, and the differences between the route selected by the user and other routes in the group are considered from multiple dimensions such as time and distance, or a clustering method such as K-Medoids clustering is used to iteratively cluster the route preference characteristics, and determine the route preference label according to the characteristics within the cluster. In this way, according to navigation requests in different time and space situations, it is possible to accurately output the route preferences that meet the user's specific time and space scenarios, thereby achieving more efficient and personalized route planning services, and achieving the effect of improving the accuracy of route recommendations and meeting user preference needs.

[0125] At the same time, when using the user preference model to produce recommended routes, there is no need for manual settings by the user. The model can learn route preferences at different times and places from the user's historical navigation behavior, solving the problem of cumbersome manual settings and difficulty in accurately grasping user preferences in existing technologies, and achieving the effect of providing users with more accurate and personalized navigation route recommendations.

[0126] Figure 3A flow chart of the route recommendation method provided for this application, such as Figure 3 As shown, the method includes:

[0127] S301: In response to a user input operation, send navigation request data input by the user to a server.

[0128] When a user performs a specific input operation on a client (e.g., a mobile phone or car navigation system), such as entering a destination in a navigation application, the client quickly captures this action. The client then packages the user's navigation request data, including possible travel time and start and end point data (i.e., detailed information about the user's specified departure and destination points), and sends it to the navigation system (server).

[0129] S302. Receive and display the recommended route sent by the server, wherein the recommended route is determined by the server based on the navigation request data and the route preference tag, the route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route, the target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are route data corresponding to the user's historical navigation request data.

[0130] When the client sends the user's navigation request data and start and end point data to the navigation system, the navigation system inputs this data into a pre-built preference recommendation model. This model, trained with extensive data, can analyze the user's navigation route preferences based on the input information, such as whether they prefer the shortest distance, fastest travel time, or avoiding toll roads. The navigation system then combines the analyzed navigation route preferences to determine a specific recommended route. Finally, the client receives the recommended route sent by the navigation system and displays it intuitively on the user interface. For example, the route is marked on the map with a brightly colored line, and relevant information such as the estimated travel time and distance is displayed, making it easy for users to view and plan their trips based on the recommended route.

[0131] The route recommendation method provided in the embodiment of the present application collects navigation request data and start and end point data input by the user and transmits them to the navigation system. The navigation system uses a preference recommendation model to analyze these data to obtain navigation route preferences, and finally feeds back the recommended route to the user and displays it. This method achieves a precise understanding of the user's personalized needs and provides the user with route recommendations that fit their multiple preferences such as time, cost, and road conditions. It effectively solves the problem that traditional navigation recommended routes are difficult to meet the complex and diverse needs of users, thereby improving the user's travel experience.

[0132] Figure 4A schematic diagram of the structure of the route recommendation device provided in this application, such as Figure 4 As shown, the route recommendation device 40 provided in this embodiment includes:

[0133] Acquisition module 401, used to obtain user's navigation request data;

[0134] Determination module 402, configured to determine a recommended route based on the user's navigation request data and a route preference tag, wherein the route preference tag is determined based on a degree of difference between a target recommended route and other recommended routes in the recommended route, and the target recommended route and other recommended routes are determined based on a degree of deviation between the recommended route and an actual travel route, wherein the recommended route and the actual travel route are route data corresponding to the user's historical navigation request data;

[0135] The sending module 403 is used to send the recommended route to the client for display in the client.

[0136] In a possible implementation, the determining module 402 is further specifically configured to:

[0137] Determining the same routes between the recommended route and the actual traveled route, and the route lengths of the same routes;

[0138] Determine the degree of deviation between the recommended route and the actual route traveled based on the route length of the same route and the actual route length of the actual route traveled;

[0139] According to the degree of deviation between the recommended route and the actual driving route, a target recommended route and other recommended routes in the recommended route are determined.

[0140] In a possible implementation, the determining module 402 is further specifically configured to:

[0141] Determine the user's preference characteristics in different preference dimensions based on the difference between the target recommended route and other recommended routes in the recommended routes;

[0142] Determine the user's route preference label based on the user's preference characteristics in different preference dimensions.

[0143] In a possible implementation, the determining module 402 is further specifically configured to:

[0144] According to the preset preference dimension, determining the characteristic extreme values of other route features in other recommended routes and the characteristic values of route features in the target recommended route features;

[0145] Determine a difference characteristic value according to characteristic extreme values of other route characteristics in other recommended routes and characteristic values of route characteristics in the target recommended route characteristics;

[0146] The user's preference characteristics under different preference dimensions are determined based on the difference feature values and the feature values of the route features in the target recommended route features.

[0147] In a possible implementation, the determining module 402 is further specifically configured to:

[0148] Determine the difference between the actual driving route and the recommended route based on the user's preference characteristics in different preference dimensions;

[0149] Determine the user's route preference label based on the difference between the actual driving route and the recommended route;

[0150] or,

[0151] Cluster the user's preference features under different preference dimensions to obtain clustering results;

[0152] Based on the clustering results, the user's route preference label is determined.

[0153] In a possible implementation, when the navigation request data includes request time data and start and end point data, and the historical navigation request data includes historical request time data and historical start and end point data, the determination module 402 is further specifically configured to:

[0154] Determine a user preference model, which is trained based on the user's route preference tags and historical navigation request data;

[0155] The navigation request data is input into the user preference model to obtain the recommended route output by the user preference model.

[0156] The route recommendation device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail in this embodiment.

[0157] Figure 5 A schematic diagram of the structure of the route recommendation device provided in this application, such as Figure 5 As shown, the route recommendation device 50 provided in this embodiment includes:

[0158] The response module 501 is used to send the navigation request data input by the user to the server in response to the user's input operation;

[0159] The display module 502 is used to receive and display the recommended route sent by the server, wherein the recommended route is determined by the server based on the navigation request data and the route preference tag, the route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route, the target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are route data corresponding to the user's historical navigation request data.

[0160] The route recommendation device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail in this embodiment.

[0161] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 6 As shown, the electronic device 60 provided in this embodiment includes: at least one processor 601 and a memory 602. Optionally, the device 60 further includes a communication component 603. The processor 601, the memory 602 and the communication component 603 are connected via a bus 604.

[0162] During the specific implementation process, at least one processor 601 executes the computer-executable instructions stored in the memory 602, so that the at least one processor 601 performs the above method.

[0163] The specific implementation process of the processor 601 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0164] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0165] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0166] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0167] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0168] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0169] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0170] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in a device as discrete components.

[0171] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, either through an interface, electrical, mechanical, or other means.

[0172] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0173] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0174] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0175] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0176] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. A route recommendation method, characterized in that: Applied to the server side, including: Get the user's navigation request data; determining a recommended route based on the user's navigation request data and a route preference tag, wherein the route preference tag is determined based on a degree of difference between a target recommended route and other recommended routes in the recommended routes, the target recommended route and the other recommended routes being determined based on a degree of deviation between the recommended route and an actual travel route, the recommended route and the actual travel route being route data corresponding to the user's historical navigation request data; The recommended route is sent to the client to be displayed in the client.

2. The method according to claim 1, characterized in that The method further comprises: determining the same route between the recommended route and the actual driving route, and the length of the same route; determining a degree of deviation between the recommended route and the actual travel route based on a route length of the same route and an actual route length of the actual travel route; The target recommended route and the other recommended routes in the recommended routes are determined according to the degree of deviation between the recommended route and the actual driving route.

3. The method according to claim 1, characterized in that The method further comprises: determining the preference characteristics of the user in different preference dimensions according to the degree of difference between the target recommended route and other recommended routes in the recommended routes; Determine the route preference label of the user according to the preference characteristics of the user in different preference dimensions.

4. The method according to claim 3, characterized in that The determining of the user's preference characteristics in different preference dimensions based on the difference between the target recommended route and other recommended routes in the recommended routes includes: Determining, based on a preset preference dimension, characteristic extreme values of other route features in the other recommended routes and characteristic values of route features in the target recommended route features; Determining a difference characteristic value based on characteristic extreme values of other route characteristics in the other recommended routes and characteristic values of route characteristics in the target recommended route characteristics; The preference characteristics of the user in different preference dimensions are determined according to the difference feature values and the feature values of the route features in the target recommended route features.

5. The method according to claim 3, characterized in that The determining of the user's route preference label according to the user's preference characteristics in different preference dimensions includes: determining the difference between the actual driving route and the recommended route based on the user's preference characteristics under different preference dimensions; determining a route preference label of the user based on the difference between the actual driving route and the recommended route; or, Clustering the preference features of the user under different preference dimensions to obtain clustering results; Determine the route preference label of the user according to the clustering result.

6. The method according to claim 1, characterized in that When the navigation request data includes request time data and start and end point data, and the historical navigation request data includes historical request time data and historical start and end point data, Determining a recommended route based on the user's navigation request data and route preference tags includes: determining a user preference model, wherein the user preference model is trained based on the user's route preference tag and the historical navigation request data; The navigation request data is input into a user preference model to obtain a recommended route output by the user preference model.

7. A route recommendation method, characterized in that: Applied to a client, the method includes: In response to a user input operation, sending the navigation request data input by the user to a server side; Receive and display the recommended route sent by the server, wherein the recommended route is determined by the server based on the navigation request data and a route preference tag, the route preference tag is determined based on the degree of difference between a target recommended route and other recommended routes in the recommended route, the target recommended route and the other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are route data corresponding to the user's historical navigation request data.

8. A route recommendation device, characterized in that: include: The response module is used to send the navigation request data input by the user to the server in response to the user's input operation; The display module is used to receive and display the recommended route sent by the server, wherein the recommended route is determined by the server based on the navigation request data and the route preference tag, the route preference tag is determined based on the degree of difference between the target recommended route and other recommended routes in the recommended route, the target recommended route and other recommended routes are determined based on the degree of deviation between the recommended route and the actual driving route, and the recommended route and the actual driving route are the route data corresponding to the user's historical navigation request data.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium / computer program product, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method according to any one of claims 1 to 7; and / or, The computer program product comprises a computer program, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images