Route recommendation method, route navigation method, and computer program product

Abstract

Embodiments of the present disclosure relate to a route recommendation method, a route navigation method and a computer program product. The route recommendation method comprises: determining a to-be-recommended route; obtaining route distinguishing feature information of the to-be-recommended route and a current main route; inputting the route distinguishing feature information as first model input features into a pre-trained acceptance rate prediction model; determining an acceptance probability of the to-be-recommended route by the acceptance rate prediction model; and determining whether to push the to-be-recommended route to a navigation object based on the acceptance probability of the to-be-recommended route. The technical solution of the embodiments of the present disclosure proposes route distinguishing feature information, accurately distinguishes the to-be-recommended route from the current main route, determines whether to push the to-be-recommended route based on the acceptance probability of the to-be-recommended route determined by the acceptance rate prediction model, and is beneficial to improving recommendation accuracy. With the passage of time, the accuracy of the acceptance probability of the to-be-recommended route determined by the acceptance rate prediction model gradually increases, and the maintenance cost is relatively low.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a route recommendation method, a route navigation method, and a computer program product. Background Technology

[0002] With the development of computer technology, navigation technology and software are increasingly used in daily life. Using navigation software, users can independently and conveniently choose routes to their destinations, improving travel convenience. However, during navigation, users may encounter situations where the current main route is no longer optimal due to sudden environmental changes. How to provide users with replanned routes at appropriate times to assist their decision-making is a problem the industry has been continuously exploring.

[0003] To address this issue, existing solutions recalculate routes in the event of sudden environmental changes. If the recalculated route meets certain rules with the current main route, it is recommended to the user. However, these rules are typically based on human experience, leading to low recommendation accuracy. Furthermore, many thresholds in the rules become invalid over time. The interrelationships between these rules result in high maintenance costs and low efficiency. Summary of the Invention

[0004] To address or at least partially address the aforementioned technical problems, this disclosure provides a route recommendation method and apparatus, a route navigation method, a medium, and a program product.

[0005] This disclosure provides a route recommendation method, including:

[0006] Determine the routes to be recommended;

[0007] Obtain the route difference feature information between the route to be recommended and the current main route;

[0008] The route differentiation feature information is used as the first model input feature and input into the pre-trained acceptance rate prediction model. The acceptance probability of the route to be recommended is determined by the acceptance rate prediction model.

[0009] Based on the acceptance probability of the proposed route, determine whether to push the proposed route to the navigation object.

[0010] This disclosure also provides a route navigation method applied to a terminal device, including:

[0011] Acquire and display the routes to be recommended to be pushed to the terminal device, wherein the routes to be recommended are determined by any of the above-mentioned route recommendation methods;

[0012] Receive recommendation acceptance data for the pushed routes to be recommended.

[0013] This disclosure also provides a route recommendation device, including:

[0014] The route determination module is used to determine the routes to be recommended.

[0015] The feature information determination module is used to obtain the route difference feature information between the route to be recommended and the current main route;

[0016] The probability determination module is used to take the route differentiation feature information as the first model input feature, input it into the pre-trained acceptance rate prediction model, and determine the acceptance probability of the route to be recommended through the acceptance rate prediction model.

[0017] The recommendation module is used to determine whether to push the recommended route to the navigation object based on the acceptance probability of the route to be recommended.

[0018] This disclosure also provides a route navigation device, including:

[0019] The route recommendation module is used to acquire and display routes to be recommended to be pushed to the terminal device, wherein the routes to be recommended are obtained by any of the above-mentioned route recommendation devices;

[0020] The results feedback module is used to receive recommendation acceptance data for the pushed routes to be recommended.

[0021] This disclosure also provides an electronic device, the electronic device comprising:

[0022] processor;

[0023] Memory used to store the processor's executable instructions;

[0024] The processor is configured to read the executable instructions from the memory and execute the instructions to implement any of the route recommendation methods or any of the route navigation methods described above.

[0025] This disclosure also provides a computer-readable storage medium storing a computer program for executing any of the above-described route recommendation methods or route navigation methods.

[0026] This disclosure also provides a computer program product for executing any of the above-described route recommendation methods or any of the above-described route navigation methods.

[0027] Compared with the prior art, the technical solution provided in this disclosure has at least the following advantages: In this disclosure, the route recommendation method can obtain route difference feature information between the determined route to be recommended and the current main route, and use this route difference feature information as the input feature of the first model, inputting it into a pre-trained acceptance rate prediction model. The acceptance rate prediction model determines the acceptance probability of the route to be recommended, and based on the acceptance probability of the route to be recommended, it determines whether to push the route to be recommended to the navigated object. The proposed route difference feature information enables accurate differentiation between the route to be recommended and the current main route, and the proposed acceptance rate prediction model enables learning of the user's behavior in accepting the route to be recommended. Therefore, determining whether to push the route to be recommended to the navigated object, and further whether to display the route to the user, based on the acceptance probability of the route to be recommended determined by the acceptance rate prediction model, helps improve the accuracy of route recommendation. Simultaneously, as time progresses, the accuracy of the acceptance probability of the route to be recommended determined by the acceptance rate prediction model gradually increases. Since this method is fully automatic and dynamically adjusted, its maintenance cost is low and its efficiency is high. Attached Figure Description

[0028] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0029] Figure 1 A flowchart illustrating a route recommendation method provided in an embodiment of this disclosure;

[0030] Figure 2 for Figure 1 The diagram shown illustrates the specific flow of step S102 in the method described.

[0031] Figure 3 A flowchart illustrating another route recommendation method provided in this embodiment of the disclosure;

[0032] Figure 4 A flowchart illustrating yet another route recommendation method provided in this disclosure embodiment;

[0033] Figure 5 A flowchart illustrating yet another route recommendation method provided in this disclosure embodiment;

[0034] Figure 6 A flowchart illustrating yet another route recommendation method provided in this disclosure embodiment;

[0035] Figure 7 A flowchart illustrating yet another route recommendation method provided in this disclosure embodiment;

[0036] Figure 8 A flowchart illustrating yet another route recommendation method provided in this disclosure embodiment;

[0037] Figure 9 A flowchart illustrating a route navigation method provided in an embodiment of this disclosure;

[0038] Figure 10 A flowchart illustrating another route navigation method provided in this embodiment of the disclosure;

[0039] Figure 11 A timing diagram of a route navigation method provided in an embodiment of this disclosure;

[0040] Figure 12 A timing diagram of another route navigation method provided in an embodiment of this disclosure;

[0041] Figure 13 This is a schematic diagram of the structure of a route recommendation device provided in an embodiment of the present disclosure;

[0042] Figure 14 This is a schematic diagram of the structure of a route navigation device provided in an embodiment of the present disclosure;

[0043] Figure 15 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0044] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0045] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0046] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies. It should be noted that the modifications "a" and "a plurality" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless explicitly indicated in the context, they should be understood as "one or more".

[0047] Existing dynamic navigation route recommendation schemes suffer from problems such as low accuracy, high maintenance costs, and numerous fallback rules.

[0048] Specifically, existing dynamic navigation technologies can use a set of rules to form a rule system to determine whether to recommend a route to the user. The specific process is as follows: During user navigation, the user's current main route is detected. When a sudden environmental change occurs on the current main route, the route is recalculated to generate a route to be recommended. The route to be recommended and the current main route are compared, and when certain rule conditions are met, the route to be recommended is pushed to the user. These rule conditions may include, for example, that the recommended route saves time compared to the current main route, that the recommended route avoids traffic accidents, or other optional rule conditions. At the same time, to avoid frequently disturbing the user, a fallback logic is set up; for example, if the user refuses to accept the route twice consecutively, it will no longer be recommended to the user.

[0049] The rules and conditions used to determine whether to recommend a route to the user are typically set based on human experience, leading to low recommendation accuracy. Furthermore, many thresholds in the rules become invalid over time. The interrelationships between rules result in high maintenance costs and low efficiency. Additionally, to avoid frequently disturbing users and causing harm, numerous fallback rules are needed to ensure the stability of the user's journey during navigation.

[0050] To address at least some of the aforementioned technical problems, this disclosure provides a route recommendation method. This method proposes route differentiation feature information to accurately distinguish the route to be recommended from the current main route, thereby improving route recommendation accuracy. It also proposes an acceptance rate prediction model capable of learning from user behavior regarding the acceptance of the recommended route. Therefore, based on the route differentiation feature information, the acceptance probability of the recommended route is determined by the acceptance rate prediction model, and the method determines whether to push the recommended route to the navigated object (i.e., the terminal device), which helps improve route recommendation accuracy. Furthermore, the accuracy of the acceptance probability determined by the acceptance rate prediction model gradually increases over time. Since this method automatically adjusts dynamically, its maintenance cost is low and its efficiency is high. Simultaneously, the recommended route is only pushed and displayed when the user is likely to accept it, eliminating the need for numerous fallback rules. This approach ensures stability during the journey while providing better route recommendations and minimizing user disruption.

[0051] The technical solutions provided in this disclosure can be applied to navigation scenarios, including navigation when users are driving, cycling, or walking, as well as navigation when users hail a taxi, ride-hailing service, or designated driver. The route recommendation method can be executed locally on the terminal device where the application or mini-program with navigation function is located, or it can be executed by a server that interacts with the terminal device, which is not limited here.

[0052] The following is combined with Figures 1-15 The present disclosure provides illustrative examples of route recommendation methods and apparatus, route navigation methods and apparatus, electronic devices, computer storage media, and computer program products provided in the embodiments of this disclosure.

[0053] Figure 1 This is a flowchart illustrating a route recommendation method provided in an embodiment of this disclosure. Figure 1 As shown, the route recommendation method may include the following steps:

[0054] S101. Determine the route to be recommended.

[0055] Among them, the route to be recommended is the route recalculated during the user's navigation process, or the route that is replanned, and is different from the current main route.

[0056] For example, the route to be recommended can be determined based on the user's current location, destination, and traffic information. This step can be implemented using any route determination method known to those skilled in the art, and is not limited herein.

[0057] S102. Obtain the route difference feature information between the route to be recommended and the current main route.

[0058] Here, the current main route is the route the user is currently navigating on. The route to be recommended differs from the current main route, forming route differentiation feature information. Therefore, the route differentiation feature information is used to characterize the differences between the route to be recommended and the current main route, i.e., the differences between the two.

[0059] For example, the route differentiation feature information may include the time difference (i.e., time advantage / disadvantage) between the recommended route and the current main route, the distance difference (i.e., distance advantage / disadvantage), the difference in the number of navigation actions (i.e., advantage / disadvantage in the number of navigation actions), the cost difference (i.e., cost advantage / disadvantage), and other route differentiation feature information that can characterize the difference between the recommended route and the current main route. The time difference may include the time saved by the recommended route compared to the current main route; the distance difference may include the additional length of the recommended route compared to the current main route; the cost difference may include the cost saved by the recommended route compared to the current main route; and the difference in the number of navigation actions may include information such as straight-line length, number of turns, number of intersections, and number of traffic lights. This is merely an illustrative example and does not constitute a limitation on the embodiments of this disclosure.

[0060] For example, route differentiation feature information can be obtained based on the difference feature between the route to be recommended and the current main route. For example, the diff function can be used to compare the route to be recommended with the current main route to obtain route differentiation feature information.

[0061] In other implementations, other methods may be used to determine the route distinguishing feature information, which are not limited here.

[0062] S103. Use the route differentiation feature information as the first model input feature, input it into the pre-trained acceptance rate prediction model, and determine the acceptance probability of the route to be recommended through the acceptance rate prediction model.

[0063] The acceptance probability of the recommended route can also be called the user acceptance rate, which is the percentage of users who accept the recommendation after it has been presented to them. For example, it can be the percentage of users who click to accept the recommendation after a pop-up window recommending the route is pushed to the navigable object and displayed on the user's side.

[0064] For example, the probability of accepting the recommended route can be between 0 and 1. The larger the value, the more likely the user is to accept the recommendation; the smaller the value, the more likely the user is to not accept the recommendation. The threshold for whether a user is inclined to accept the recommendation can be set based on the needs of the route recommendation method, and is not limited here.

[0065] The acceptance rate prediction model is a model used to learn user acceptance behavior. It can determine the acceptance probability of the route to be recommended based on route differentiation feature information. In subsequent steps, the route to be recommended can only be recommended to the user when the user is inclined to accept the recommendation, thereby improving the accuracy of route recommendation.

[0066] The route differentiation feature information can be used as input to the acceptance rate prediction model, and the acceptance probability can be used as output. Based on this, upon receiving the route differentiation feature information, the acceptance rate prediction model can output the corresponding acceptance probability of the recommended route, so that subsequent steps can determine whether to push the recommended route to the navigated object, and further, whether to display the recommended route to the user, as shown in S104. For example, the acceptance rate prediction model can be a linear model, an ensemble learning-related model, a neural network model, a deep learning-related model, or other types of probabilistic models; these are not elaborated upon or limited here.

[0067] The pre-training of the acceptance rate prediction model can be based on a large amount of historical data. This historical data includes route differentiation features and the probability of a user accepting the route recommendation. It can further include route feature information of the route to be recommended, route feature information of the current main route, and other factors influencing whether a user accepts the route recommendation, as detailed later. The pre-training method for the acceptance rate prediction model can be any model pre-training method known to those skilled in the art, and will not be elaborated upon or limited here.

[0068] Therefore, in this route recommendation method, the acceptance rate prediction model is fine-tuned using real-time data over time, thereby further improving the prediction accuracy of the acceptance rate prediction model and thus enhancing the accuracy of route recommendation.

[0069] S104. Based on the acceptance probability of the route to be recommended, determine whether to push the route to be recommended to the navigation object.

[0070] In this case, when the probability of accepting the recommended route is low, users tend not to accept the recommendation. In this case, it is not necessary to show the recommended route to the user to avoid disturbing the user. Correspondingly, it is not necessary to push the recommended route to the navigated object. When the probability of accepting the recommended route is high, users tend to accept the recommendation. In this case, the recommended route can be shown to the user. Correspondingly, the recommended route can be pushed to the navigated object, so that the user can choose a route again.

[0071] Therefore, based on the acceptance probability of the recommended route, that is, based on whether the user is inclined to accept the recommendation, it is determined whether to show the recommended route to the user. Correspondingly, it is determined whether to push the recommended route to the navigable object, which helps to improve the accuracy of route recommendation. At the same time, it can avoid frequently disturbing the user and ensure the stability of the user's journey.

[0072] In the route recommendation method provided in this embodiment, the acceptance probability of the route to be recommended is determined by an acceptance rate prediction model based on route differentiation feature information, and further, it is determined whether to push the recommended route to the navigable object, which helps to improve the accuracy of route recommendation. Furthermore, as time progresses, the acceptance rate prediction model is fine-tuned based on the input features of the first model, and the accuracy of its determination of the acceptance probability of the recommended route gradually increases. Since this method performs fully automatic dynamic adjustments, its maintenance cost is low and its efficiency is high. Simultaneously, the recommended route is only pushed and displayed when the user is likely to accept it, eliminating the need for numerous fallback rules. This approach ensures stability during the journey while providing better route recommendations and minimizing user disruption.

[0073] In some embodiments, Figure 2 for Figure 1 The illustrated flowchart shows the specific process of S102 in the method shown. Figure 1 Based on, refer to Figure 2 S102 may include the following steps:

[0074] S201. Obtain the route feature information of the route to be recommended, as well as the route feature information of the current main route.

[0075] The route feature information for the route to be recommended may include the time, distance, number of navigation actions, and cost of the route to be recommended; the route feature information for the current main route may include the time, distance, number of navigation actions, and cost of the current main route.

[0076] For example, in this step, the route feature information of the route to be recommended and the route feature information of the current main route can be obtained respectively. The method of obtaining them can be any method known to those skilled in the art, which will not be elaborated or limited here.

[0077] S202. Based on the route feature information of the route to be recommended and the route feature information of the current main route, determine the route difference feature information between the route to be recommended and the current main route.

[0078] Among them, at least one of the route feature information of the route to be recommended and the route feature information of the current main route differs in terms of time, distance, number of navigation actions, and cost. By comparing the route feature information of the route to be recommended and the route feature information of the current main route, the route difference feature information between the route to be recommended and the current main route can be obtained.

[0079] For example, in conjunction with the above, there may be at least one difference between the route to be recommended and the current main route, such as time difference, distance difference, number of navigation actions difference, and cost difference.

[0080] This allows us to obtain route difference information between the route to be recommended and the current main route.

[0081] In some embodiments, Figure 1 and Figure 2 Based on this, the first model input features further include: route feature information of the route to be recommended and route feature information of the current main route. Correspondingly, in the route recommendation method provided in this embodiment, Figure 1 S103 in the text can be replaced with:

[0082] The route features of the route to be recommended, the route features of the current main route, and the route differentiation features are used as the input features of the first model. These are then input into the pre-trained acceptance rate prediction model, which determines the acceptance probability of the route to be recommended.

[0083] In addition to using route differentiation features as input to the acceptance rate prediction model, the route features of the route to be recommended and the route features of the current main route can also be used as input to the acceptance rate prediction model. This is equivalent to predicting the acceptance probability of the route to be recommended based on the above three aspects of information, which helps to improve the accuracy of the acceptance probability of the route to be recommended, thereby improving the route recommendation accuracy.

[0084] In some embodiments, Figure 3 This is a flowchart illustrating another route recommendation method provided in an embodiment of this disclosure. Figure 1 Based on, refer to Figure 3 S103 may also include:

[0085] S1021. Obtain at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information.

[0086] The recommended route and the current main route may have intersections. Intersection feature information is used to characterize the differences between the recommended route and the current main route at the intersection. For example, the intersection feature information for the recommended route might be a left turn, while the intersection feature information for the current main route might be a right turn; or the intersection feature information for the recommended route might be to continue straight, while the intersection feature information for the current main route might be a right turn. This can be determined based on the specific routes of the recommended route and the current main route, and is not limited here.

[0087] The user's current location information can be their real-time location information, such as whether they are at a traffic light intersection, near a school or residential area (limited by whether the actual distance to the school or residential area is within a preset distance range, which can be 300 meters, 500 meters, or other distance values, and is not limited here), or at a bridgehead or on a bridge, or other real-time location information, which is not limited here. The user's current location information can be obtained based on a positioning system and transmitted to the route recommendation device.

[0088] The user's current speed information can be the user's real-time speed information. For example, in a scenario where the user is driving autonomously, the user's current speed information may include the speed per hour. The user's current speed information can be obtained based on a speed monitoring system and transmitted to the route recommendation device.

[0089] Among them, user profile information can be used to characterize users' historical preferences. For example, whether a user prefers short time, short distance, few navigation actions, low cost, or other historical preferences can be obtained based on historical user behavior statistics and transmitted to the route recommendation device.

[0090] Based on this Figure 1 S103 in Figure 3 Specifically, it can be S1031, as follows:

[0091] S1031. Take at least one of the following as the first model input features: route differentiation feature information, intersection feature information of the route to be recommended, intersection feature information of the current main route, user current location information, user current speed information, and user profile information. Input these features into the pre-trained acceptance rate prediction model and determine the acceptance probability of the route to be recommended through the acceptance rate prediction model.

[0092] In this way, the acceptance probability of the route to be recommended can be determined by an acceptance rate prediction model based on route differentiation feature information, intersection feature information of the route to be recommended, intersection feature information of the current main route, user current location information, user current speed information, and user profile information.

[0093] In addition to using route differentiation features as input to the acceptance rate prediction model, at least one of the following can also be used as input: intersection features of the route to be recommended, intersection features of the current main route, user's current location, user's current speed, and user profile. This is equivalent to predicting the acceptance probability of the route to be recommended based on at least two different aspects of information, which helps to improve the accuracy of the acceptance probability of the route to be recommended, thereby improving the accuracy of route recommendation.

[0094] In other embodiments, the above-mentioned Figure 2 The corresponding solution and Figure 3The corresponding solutions are combined to further improve the accuracy of the acceptance rate prediction model. For example, the first model input features may include route feature information of the route to be recommended, route feature information of the current main route, and route differentiation feature information, as well as at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information. This increases the total amount of data for further adjustment of the acceptance rate prediction model, which is beneficial to improving the accuracy of the acceptance rate prediction model.

[0095] In some embodiments, Figure 4 This is a flowchart illustrating yet another route recommendation method provided in an embodiment of this disclosure. Figure 1 Based on, refer to Figure 4 In this route recommendation method, after S104, it may also include:

[0096] S105. Obtain the recommendation acceptance result data of the navigated object for the pushed recommended route.

[0097] Among them, the user feedback recommendation acceptance data is used to characterize whether the user accepts the recommendation, which can include accepting the recommendation and not accepting the recommendation.

[0098] In this step, after pushing the recommended route to the navigated object and further displaying the recommended route to the user, the navigated object receives the recommendation acceptance result data fed back by the user. Furthermore, it obtains the recommendation acceptance result data fed back by the navigated object for the pushed recommended route, providing a data foundation for subsequent training of the acceptance rate prediction model.

[0099] For example, both the user's feedback on the recommendation acceptance result and the corresponding feedback on the recommendation acceptance result of the navigated object regarding the pushed route to be recommended can be represented numerically. For instance, when a user accepts the recommendation, the user's feedback on the recommendation acceptance result can be 1, and correspondingly, the navigated object's feedback on the recommendation acceptance result of the pushed route to be recommended is 1; conversely, when a user does not accept the recommendation, the user's feedback on the recommendation acceptance result can be 0, and correspondingly, the navigated object's feedback on the recommendation acceptance result of the pushed route to be recommended is 0.

[0100] S106. Use the recommendation acceptance result data as the output feature of the first model, input it into the acceptance rate prediction model, and train the acceptance rate prediction model.

[0101] After obtaining the recommendation acceptance results data from the navigation object for the pushed recommended routes, this data can be used as the output feature of the first model to train the acceptance rate prediction model. The training method can be incremental optimization iteration to optimize the acceptance rate prediction model, increase the acceptance probability of the recommended routes determined by the acceptance rate prediction model, and thus improve the recommendation accuracy.

[0102] The training samples for the acceptance rate prediction model may include user feedback on recommended acceptance results and route differentiation features, as well as route feature information of the route to be recommended, the current main route, and / or at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information. The acceptance rate prediction model is trained using these training samples to learn user behavior, thereby enabling a more accurate determination of the acceptance probability of the recommended route.

[0103] The timing for training the acceptance rate prediction model can be as follows: after a user rejects a suggested route for the first time, collect samples of this rejection, and use these samples to train the acceptance rate prediction model after all trips for the day have ended, for example, through iterative optimization. The optimized acceptance rate prediction model can then be applied to the route recommendation method the following day.

[0104] In other implementations, other time intervals can be set, such as 2 days, 3 days, etc., to periodically train the acceptance rate prediction model; or the acceptance rate prediction model can be trained based on user triggers, or based on other timings, which will not be elaborated or limited here.

[0105] In some embodiments, Figure 5 This is a flowchart illustrating yet another route recommendation method provided in an embodiment of this disclosure. Figure 1 Based on, refer to Figure 5 The route recommendation method may also include the following before S104:

[0106] S1032. Use the route differentiation feature information as the input feature of the second model, input it into the pre-trained yaw probability prediction model, and determine the yaw probability of the route to be recommended through the yaw probability prediction model.

[0107] Among them, deviation refers to the situation where the user's actual route is inconsistent with the navigation route selected. In this case, the navigation route is the one that the user accepted as a recommendation, i.e., the route to be recommended.

[0108] In practical applications, if a user accepts a recommendation and does not deviate from the intended route, then the user's behavior is as expected, meaning they are following the accepted recommended route. Correspondingly, it is necessary to recommend a replanned route to the user, i.e., it is necessary to push the recommended route to the navigated object. However, if the user deviates from the intended route after accepting the recommendation, then the user's behavior is not as expected, i.e., the user did not follow the recommended route. For example, this could be due to a user's erroneous action in accepting the recommendation but subsequently not following it. Correspondingly, it is unnecessary to recommend a replanned route to the user, i.e., it is unnecessary to push the recommended route to the navigated object.

[0109] For example, the deviation probability of the recommended route can be between 0 and 1. The larger the value, the higher the probability of the user deviating from the route; the smaller the value, the lower the probability of the user deviating from the route. The threshold for whether a user may deviate from the route can be set based on the requirements of the route recommendation method, and is not limited here.

[0110] Among them, the yaw probability prediction model is a model used to learn the user's yaw behavior. It can determine the yaw probability of the route to be recommended based on route distinguishing feature information. In subsequent steps, routes can be recommended to users only when the probability of user yaw is small, thereby improving the accuracy of route recommendation.

[0111] The yaw probability prediction model takes route differentiation feature information as input and outputs yaw probability. Based on this, upon receiving route differentiation feature information, the yaw probability prediction model can output the yaw probability of the corresponding recommended route. This information is then used in subsequent steps to determine whether to push the recommended route to the navigable object and, more importantly, whether to display the recommended route to the user (see S1041).

[0112] The pre-training of the yaw probability prediction model can be based on a large amount of historical data. This historical data includes route distinguishing features and corresponding yaw probabilities. Further, it may include route feature information of the route to be recommended, route feature information of the current main route, and other factors that influence whether the user accepts the route recommendation, as detailed later. The pre-training method for the yaw probability prediction model can be any model pre-training method known to those skilled in the art, and will not be elaborated upon or limited here.

[0113] Therefore, in this route recommendation method, the yaw probability prediction model is fine-tuned using real-time data over time, thereby further improving the prediction accuracy of the yaw probability prediction model and thus enhancing the accuracy of the route recommendation method.

[0114] Based on this Figure 1 S104 in the text can be replaced with Figure 5 S1401 in the example is as follows:

[0115] S1041. Based on the acceptance probability of the route to be recommended and the deviation probability of the route to be recommended, determine whether to push the route to be recommended to the navigation object.

[0116] Based on the above, the acceptance probability of a recommended route can represent the percentage of users who accept the recommendation, while the deviation probability of a recommended route can represent the likelihood of users deviating from the recommended route. Therefore, when the percentage of users who accept the recommendation is high and the likelihood of users deviating from the recommended route is low, it indicates that users have a high acceptance and practicality of the recommended route, and in this case, it is determined to push and display the recommended route; otherwise, it is determined not to push the recommended route.

[0117] For example, when both the acceptance probability and the deviation probability of the recommended route are between 0 and 1, if the acceptance probability of the recommended route is greater than 50% and the deviation probability of the recommended route is less than 50%, then the recommended route will be pushed and displayed; otherwise, the recommended route will not be pushed.

[0118] In other implementations, other threshold conditions can be used to determine whether to push recommended routes. These conditions can be set based on the requirements of the route recommendation method and will not be elaborated or limited here.

[0119] Therefore, the route recommendation method provided in this embodiment determines whether to push the route to the navigable object, i.e. whether to show the user the recommended route, by learning the user's acceptance behavior and deviation behavior after accepting the recommendation. This can reduce the interference caused by mistakenly accepting the recommended route. Moreover, the method can be fully automated, and there is no threshold failure problem. It has low maintenance cost and high efficiency.

[0120] In some embodiments, Figure 5 Building upon this, the second model's input features further include: route feature information of the route to be recommended and route feature information of the current main route. Therefore, Figure 5 S1032 can be replaced with:

[0121] The route feature information of the route to be recommended, the route feature information of the current main route, and the route difference feature information are used as the input features of the second model. These are then input into the pre-trained yaw probability prediction model, which determines the yaw probability of the route to be recommended.

[0122] In addition to using route differentiation feature information as input to the yaw probability prediction model, the route feature information of the route to be recommended and the route feature information of the current main route can also be used as input to the yaw probability prediction model. This is equivalent to predicting the yaw probability of the route to be recommended based on the above three aspects of information, which helps to improve the accuracy of the yaw probability of the route to be recommended, thereby improving the accuracy of route recommendation.

[0123] In some embodiments, Figure 6 This is a flowchart illustrating yet another route recommendation method provided in an embodiment of this disclosure. Figure 5 Based on, refer to Figure 6 The route recommendation method may also include the following before S1032:

[0124] S1022. Obtain at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information.

[0125] This step and Figure 3 The same applies to S1021, which can be understood by referring to the explanation of S1021 above, and will not be repeated here.

[0126] Based on this, the second model input features further include at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information. Therefore, Figure 5 S1032 in the text can be replaced with Figure 6 S1033 in the example is as follows:

[0127] S1033. Take at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user current location information, user current speed information, and user profile information, as well as route difference feature information, as input features of the second model, input them into the pre-trained yaw probability prediction model, and determine the yaw probability of the route to be recommended through the yaw probability prediction model.

[0128] In this way, the yaw probability of the route to be recommended is determined by a yaw probability prediction model based on at least one of the intersection feature information of the route to be recommended, the intersection feature information of the current main route, the user's current location information, the user's current speed information, and the user profile information, as well as the route differentiation feature information.

[0129] In addition to using route differentiation features as input to the yaw probability prediction model, at least one of the following can also be used as input: intersection features of the route to be recommended, intersection features of the current main route, user's current location, user's current speed, and user profile. This is equivalent to predicting the yaw probability of the route to be recommended based on at least two different aspects of information, which helps improve the accuracy of the yaw probability of the route to be recommended and thus improves the route recommendation accuracy.

[0130] In other embodiments, the input features used to train the yaw probability prediction model can be integrated to further improve the accuracy of the acceptance rate prediction model. For example, the second model input features may include route feature information of the route to be recommended, route feature information of the current main route, and route differentiation feature information, as well as at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information. This increases the total amount of data for further adjustment of the yaw probability prediction model, which is beneficial for improving the accuracy of the yaw probability prediction model.

[0131] In some embodiments, Figure 7 This is a flowchart illustrating yet another route recommendation method provided in an embodiment of this disclosure. Figure 5 Based on, refer to Figure 7 In this route recommendation method, the route after S may also include:

[0132] S107. Obtain the yaw result data of the navigable object for the accepted route to be recommended.

[0133] Among them, the yaw result data is used to characterize whether the user deviates from the recommended route, and can include yaw and no yaw.

[0134] In this step, after the user accepts the recommended route, the actual route traveled by the user is obtained and compared with the recommended route to determine whether the user has deviated from the recommended route, providing a data foundation for subsequent training of the deviation probability prediction model. For example, if the user's actual route matches the recommended route, there is no deviation; if they do not match, there is a deviation.

[0135] For example, the yaw result data can be represented numerically. For instance, if the user veers off course after accepting the recommended route, the yaw result data can be 1; if the user does not veer off course after accepting the recommended route, the yaw result data can be 0.

[0136] S108. Use the yaw result data as the output feature of the second model, input it into the yaw probability prediction model, and train the yaw probability prediction model.

[0137] After obtaining the yaw result data, it can be used as the output feature of the second model to train the yaw probability prediction model. The training method can be incremental optimization iteration to optimize the yaw probability prediction model, improve the yaw probability of the route to be recommended determined by the yaw probability prediction model, and thus improve the route recommendation accuracy.

[0138] The training samples for the yaw probability prediction model may include yaw result data and route differentiation features, as well as route feature information of the route to be recommended, route feature information of the current main route, and / or at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information. The yaw probability prediction model is trained using these training samples to learn user behavior, thereby enabling it to more accurately determine the yaw probability of the route to be recommended.

[0139] Based on the above, the yaw probability prediction model can be trained synchronously with the acceptance rate prediction model, that is, the training timing of the two can be consistent; or, the training timing of the two can be asynchronous but set in association, or they can be set independently, which can be set based on the needs of the route recommendation method. This will not be elaborated or limited here.

[0140] In some embodiments, Figure 8 This is a flowchart illustrating yet another route recommendation method provided in an embodiment of this disclosure. Figure 1 Based on, refer to Figure 8 Prior to S101, it may also include:

[0141] S1001. Obtain the traffic information of the current main route.

[0142] It can monitor traffic conditions on the current main route in real time or periodically at preset time intervals. The preset time interval can be selected as 1 minute, 2 minutes, or other durations, and is not limited here.

[0143] For example, the traffic information for the current main route may include information on the number of vehicles on the route, congestion level, emergencies, and other information related to navigation.

[0144] For example, the current main route's road condition information can be obtained by: obtaining it based on vehicle network monitoring data, or by any other method known to those skilled in the art.

[0145] S1002. When the road condition information of the current main route determines that the conditions for replanning are met, determine the route to be recommended.

[0146] The rerouting condition can be a sudden environmental change on the current main route that renders it unsuitable for the user's navigation needs. For example, rerouting conditions could include: sudden traffic congestion, a sudden car accident, a sudden ground collapse, or other unforeseen events ahead.

[0147] In this step, based on the current main route's traffic information, a judgment is made, and if it is determined that the conditions for replanning are met, the route is recalculated to determine the route to be recommended.

[0148] In this way, navigation routes can be selectively switched based on traffic information. That is, dynamic route recommendations are made as time and environment change dynamically during the user's navigation, thus achieving dynamic navigation that meets the user's needs.

[0149] It should be noted that the above Figures 1-8 The same steps can be understood by referring to the explanations in other corresponding locations, and will not be repeated here.

[0150] Based on the above embodiments, this disclosure also provides a route navigation method, which can be applied to a terminal device (i.e., the object being navigated). Exemplarily, the terminal device can be a smart terminal, an in-vehicle terminal, or other terminal devices, and is not limited thereto.

[0151] For example, Figure 9 This is a flowchart illustrating a route navigation method provided in an embodiment of this disclosure. (Refer to...) Figure 9 The route navigation method may include:

[0152] S301. Obtain and display the routes to be recommended to the terminal devices.

[0153] The route to be recommended is determined by any of the route recommendation methods described in the above embodiments. Therefore, the recommended route has high accuracy, which is beneficial for meeting the user's dynamic navigation needs and does not frequently disturb the user.

[0154] For example, the route recommendation method can be executed locally on the terminal device or on a server that interacts with the terminal device; correspondingly, it can be based on routes obtained locally on the terminal device or routes generated in the cloud that the terminal device obtains, as will be discussed later. Figure 10 Note: This is not a limitation.

[0155] For example, the route to be recommended can be displayed using a device or device component with display function, such as a liquid crystal display, but this is not limited to that.

[0156] S302, Receive recommendation acceptance result data for the pushed routes to be recommended.

[0157] In this process, after the terminal device receives the pushed recommended routes and displays them to the user, the user interacts with the terminal device to accept or reject the recommendations. Correspondingly, the terminal device responds to the user's interaction by generating recommendation acceptance result data, that is, generating recommendation acceptance result data for the pushed recommended routes, thereby enabling the terminal device to receive recommendation acceptance result data for the pushed recommended routes.

[0158] In this way, navigation routes are recommended through terminal devices, thus achieving dynamic navigation. At the same time, user feedback on the acceptance of recommendations is generated, and the feedback on the acceptance of recommendations from the navigated objects for the pushed routes is received. This provides sample data for training the acceptance rate prediction model, which is conducive to optimizing the acceptance rate prediction model and thus improving the recommendation accuracy.

[0159] In some embodiments, Figure 10 This is a flowchart illustrating another route navigation method provided in an embodiment of this disclosure, showing the route navigation method executed on a server using a route recommendation method. Figure 9 Based on, refer to Figure 10 In this route navigation method, S301 may specifically include:

[0160] S3011, Receive the recommended route from the server.

[0161] S3012, Show the user the routes to be recommended.

[0162] The route to be recommended is determined by the route recommendation method executed on the server.

[0163] For example, the server executes a route recommendation method, determines the recommended route to be pushed to the navigated object, and then sends the recommended route to the terminal device; correspondingly, the terminal device receives the recommended route sent by the server. This reduces the data processing flow and workload of the terminal device, helping to ensure a faster response speed.

[0164] In some embodiments, Figure 11 This is a timing diagram of a route navigation method provided in an embodiment of this disclosure. Figure 9 Based on, refer to Figure 11 The route navigation method may also include:

[0165] Receive recommendation acceptance data from user feedback;

[0166] Send the recommended acceptance result data to the server.

[0167] Among them, the terminal device interacts with the user and the server to realize the transmission of recommendation acceptance result data.

[0168] Specifically, after the terminal device displays the recommended route to the user, it can generate recommendation acceptance result data based on the user's actions, and correspondingly generate recommendation acceptance result data for the already pushed recommended route, and send the recommendation acceptance result data to the server. Correspondingly, the server receives the user's recommendation acceptance result data, trains the acceptance rate prediction model, optimizes the acceptance rate prediction model, thus forming a quality closed loop, improving recommendation accuracy, and meeting the user's dynamic navigation needs.

[0169] In some embodiments, Figure 12 A timing diagram for another route navigation method provided in an embodiment of this disclosure. Figure 11 Based on, refer to Figure 12 The route navigation method may also include:

[0170] Obtain yaw result data after the user accepts the recommended route;

[0171] Send yaw result data to the server.

[0172] Among them, the terminal device interacts with the user and the server, and can also transmit yaw result data.

[0173] Specifically, after the terminal device displays a recommended route to the user and the user accepts the recommendation, the terminal device can monitor the user's actual travel route to compare it with the recommended route, obtain yaw result data, and send the yaw result data to the server. Correspondingly, the server receives the yaw result data, trains the yaw probability prediction model, optimizes the yaw probability prediction model, thereby forming a quality closed loop, improving recommendation accuracy, and thus meeting the user's dynamic navigation needs.

[0174] This disclosure also provides a route recommendation device, which can be used to execute the steps of any of the above-described route recommendation methods to achieve the corresponding effects.

[0175] For example, Figure 13 This is a schematic diagram of a route recommendation device provided in an embodiment of the present disclosure. (Refer to...) Figure 13 The route recommendation device 410 may include:

[0176] Route determination module 411 is used to determine the route to be recommended;

[0177] The feature information determination module 412 is used to obtain the route difference feature information between the route to be recommended and the current main route;

[0178] The probability determination module 413 is used to take the route differentiation feature information as the first model input feature, input it into the pre-trained acceptance rate prediction model, and determine the acceptance probability of the route to be recommended through the acceptance rate prediction model.

[0179] The recommendation module 414 is used to determine whether to push the recommended route to the navigation object based on the acceptance probability of the route to be recommended.

[0180] In the route recommendation device provided in this embodiment, through the synergistic effect of the above-mentioned modules, route differentiation feature information between the determined route to be recommended and the current main route can be obtained. This route differentiation feature information is used as the input feature of the first model and input into a pre-trained acceptance rate prediction model. The acceptance rate prediction model determines the acceptance probability of the route to be recommended, and based on the acceptance probability of the route to be recommended, it is determined whether to push the route to be recommended to the navigated object. The proposed route differentiation feature information enables accurate differentiation between the route to be recommended and the current main route, and the proposed acceptance rate prediction model enables learning of the user's behavior in accepting the route to be recommended. Therefore, determining whether to push the route to be recommended to the navigated object, and further, whether to display the route to the user, based on the acceptance probability of the route to be recommended determined by the acceptance rate prediction model, helps improve the accuracy of route recommendation. Furthermore, as time goes on, the accuracy of the acceptance probability of the route to be recommended determined by the acceptance rate prediction model gradually increases. Since this method is fully automatic and dynamically adjusted, its maintenance cost is low and its efficiency is high.

[0181] In some embodiments, the feature information determination module 412 is specifically used for:

[0182] Obtain the route feature information of the route to be recommended, as well as the route feature information of the current main route;

[0183] Based on the route feature information of the route to be recommended and the route feature information of the current main route, the route difference feature information between the route to be recommended and the current main route is determined.

[0184] In some embodiments, the probability determination module 413 is specifically used for:

[0185] The route features of the route to be recommended, the route features of the current main route, and the route differentiation features are used as the input features of the first model. These are then input into the pre-trained acceptance rate prediction model, which determines the acceptance probability of the route to be recommended.

[0186] In some embodiments, the route recommendation device may further include:

[0187] The auxiliary information acquisition module is used to acquire at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information;

[0188] The probability determination module 413 is specifically used for:

[0189] The first model input features include route differentiation features, intersection features of the route to be recommended, intersection features of the current main route, user's current location, user's current speed, and user profile. These features are then input into a pre-trained acceptance rate prediction model, which determines the acceptance probability of the route to be recommended.

[0190] In some embodiments, the route recommendation device further includes:

[0191] The first result acquisition module is used to acquire the recommendation acceptance result data of the navigation object for the pushed recommended route;

[0192] The first training module is used to train the acceptance rate prediction model by taking the recommendation acceptance results data from user feedback as the output features of the first model and inputting them into the acceptance rate prediction model.

[0193] In some embodiments, the route recommendation device further includes:

[0194] The yaw probability determination module is used to take route differentiation feature information as input features of the second model, input it into the pre-trained yaw probability prediction model, and determine the yaw probability of the route to be recommended through the yaw probability prediction model.

[0195] Recommended module 414 is specifically used for:

[0196] Based on the acceptance probability of the recommended route and the deviation probability of the recommended route, determine whether to push the recommended route to the navigation object.

[0197] In some embodiments, the yaw probability determination module is specifically used for:

[0198] The route feature information of the route to be recommended, the route feature information of the current main route, and the route difference feature information are used as the input features of the second model. These are then input into the pre-trained yaw probability prediction model, which determines the yaw probability of the route to be recommended.

[0199] In some embodiments, the route recommendation device further includes:

[0200] The auxiliary information acquisition module acquires at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information;

[0201] The yaw probability determination module is specifically used for:

[0202] The second model input features include at least one of the following: intersection features of the route to be recommended, intersection features of the current main route, user's current location, user's current speed, and user profile information, as well as route differentiation features. These features are then input into a pre-trained yaw probability prediction model, which determines the yaw probability of the route to be recommended.

[0203] In some embodiments, the route recommendation device further includes:

[0204] The second result acquisition module is used to acquire the yaw result data of the navigated object for the accepted route to be recommended;

[0205] The second training module is used to input the yaw result data as the output features of the second model into the yaw probability prediction model to train the yaw probability prediction model.

[0206] In some embodiments, the route recommendation device further includes:

[0207] The traffic information acquisition module is used to obtain traffic information for the current main route;

[0208] The condition judgment module is used to determine the route to be recommended when the road condition information of the current main route meets the conditions for replanning.

[0209] The route recommendation apparatus disclosed in the above embodiments can implement the route recommendation method disclosed in the above method embodiments and has the same or corresponding beneficial effects. To avoid repetition, it will not be described again here.

[0210] This disclosure also provides a route navigation device for performing the steps of any of the above-described route navigation methods to achieve the corresponding effects.

[0211] In some embodiments, Figure 14 This is a schematic diagram of the structure of a route navigation device provided in an embodiment of this disclosure. (Refer to...) Figure 14 The route navigation device 420 may include:

[0212] The route recommendation module 421 is used to acquire and display the route to be recommended to be pushed to the terminal device. The route to be recommended is obtained by any of the above-mentioned route recommendation devices 410.

[0213] The result feedback module 422 is used to receive the recommendation acceptance result data for the pushed routes to be recommended.

[0214] In some embodiments, the route navigation device further includes:

[0215] The route receiving module is used to receive the route to be recommended from the server before displaying the route to be recommended to the user. The route to be recommended is determined by the route recommendation method executed on the server.

[0216] In some embodiments, the route navigation device further includes:

[0217] The result sending module is used to send the recommended acceptance result data to the server.

[0218] In some embodiments, the route navigation device further includes:

[0219] The yaw result feedback module is used to obtain yaw result data for accepted routes to be recommended;

[0220] The yaw result sending module is used to send the yaw result data to the server.

[0221] The route navigation device disclosed in the above embodiments can realize the route navigation method disclosed in the above method embodiments and has the same or corresponding beneficial effects. To avoid repetition, it will not be described again here.

[0222] Figure 15 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. The route recommendation or route navigation scheme described above in this embodiment can be implemented on a cloud server or a local host. Therefore, the electronic device can be a cloud server, a local host, or a client device or terminal device as described above.

[0223] The following detailed reference illustrates a structural diagram suitable for implementing the electronic device 500 in the embodiments of this disclosure. The electronic device 500 in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The illustrated electronic device is merely an example and should not impose any limitation on the functionality or scope of the embodiments of this disclosure.

[0224] like Figure 15As shown, electronic device 500 may include a processor (e.g., central processing unit, graphics processing unit, etc., also referred to as a processing device) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503. RAM 503 also stores various programs and data required for the operation of electronic device 500. Processor 501, ROM 502, and RAM 503 are interconnected via bus 504. Input / output (I / O) interface 505 is also connected to bus 504. Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although an electronic device 500 with various devices is shown, it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or included alternatively.

[0225] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 509, or installed from storage device 508, or installed from ROM 502. When the computer program is executed by processing device 501, the processes defined in the route recommendation method or route navigation method of embodiments of this disclosure are performed to achieve the corresponding functions.

[0226] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0227] In some implementations, the client and server can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol), and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0228] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0229] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0230] Determine the routes to be recommended;

[0231] Obtain the route difference features between the route to be recommended and the current main route;

[0232] The route differentiation feature information is used as the first model input feature and input into the pre-trained acceptance rate prediction model. The acceptance probability of the route to be recommended is determined by the acceptance rate prediction model.

[0233] Based on the acceptance probability of the recommended route, determine whether to push the recommended route to the navigation object.

[0234] Alternatively, the aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0235] Obtain and display the push recommendation route to the terminal device, the recommendation route being determined by any of the above route recommendation methods;

[0236] Receive recommendation acceptance data for the pushed routes to be recommended.

[0237] Alternatively, the aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to perform any of the aforementioned route recommendation or route navigation methods. To avoid repetition, further details are omitted here.

[0238] The computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0239] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0240] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0241] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0242] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0243] According to one or more embodiments of the present disclosure, the present disclosure provides a computer-readable storage medium storing a computer program for performing any of the route recommendation methods provided in the embodiments of the present disclosure, or for performing any of the route navigation methods provided in the embodiments of the present disclosure.

[0244] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0245] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0246] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A route recommendation method, comprising: Determine the routes to be recommended; Obtain the route difference feature information between the route to be recommended and the current main route; The route differentiation features include at least one of the following: time difference, distance difference, number of navigation actions difference, and cost difference. The route differentiation feature information is used as the input feature of the first model and input into the pre-trained acceptance rate prediction model. The acceptance probability of the route to be recommended is determined by the acceptance rate prediction model. The acceptance probability of the route to be recommended is the percentage of users who accept the recommendation after the route to be recommended is recommended to them. Based on the acceptance probability of the proposed route, determine whether to push the proposed route to the navigation object.

2. The method according to claim 1, wherein, The step of obtaining the route difference feature information between the route to be recommended and the current main route includes: Obtain the route feature information of the route to be recommended, and the route feature information of the current main route; Based on the route feature information of the route to be recommended and the route feature information of the current main route, the route difference feature information between the route to be recommended and the current main route is determined.

3. The method according to claim 2, wherein, The first model input features also include: route feature information of the route to be recommended and route feature information of the current main route.

4. The method according to claim 2, further comprising: Obtain at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information; The first model input features also include at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user current location information, user current speed information, and user profile information.

5. The method according to any one of claims 1-4, further comprising: Obtain the recommendation acceptance data of the navigated object for the pushed recommended route; The recommended acceptance result data is used as the output feature of the first model and input into the acceptance rate prediction model to train the acceptance rate prediction model.

6. The method according to claim 2, further comprising: The route feature information of the route to be recommended, the route feature information of the current main route, and the route difference feature information are used as the input features of the second model and input into the pre-trained yaw probability prediction model. The yaw probability of the route to be recommended is determined by the yaw probability prediction model. The step of determining whether to push the recommended route to the navigation object based on the acceptance probability of the recommended route includes: Based on the acceptance probability of the proposed route and the deviation probability of the proposed route, it is determined whether to push the proposed route to the navigation object.

7. The method according to claim 6, further comprising: Obtain at least one of the following: intersection feature information of the route to be recommended, intersection feature information of the current main route, user's current location information, user's current speed information, and user profile information; The second model input features also include at least one of the following: intersection feature information based on the route to be recommended, intersection feature information based on the current main route, user current location information, user current speed information, and user profile information.

8. The method according to claim 6, further comprising: Obtain the yaw result data of the navigated object for the route to be recommended that has been accepted; The yaw result data is used as the output feature of the second model and input into the yaw probability prediction model to train the yaw probability prediction model.

9. The method according to claim 4 or 6, wherein, The route differentiation features include at least one of the following: time difference, distance difference, number of navigation actions difference, and cost difference between the recommended route and the current main route; The route feature information of the route to be recommended includes at least one of the following: time, distance, number of navigation actions, and cost. The route feature information of the current main route includes at least one of the following: time, distance, number of navigation actions, and cost of the current main route; The intersection feature information of the route to be recommended includes one of the following in the route to be recommended: turning left, turning right, or continuing straight at the intersection. The intersection feature information of the current main route includes one of the following options at the intersection location: continue straight, turn left, or turn right: The user's current location information includes at least one of the following: whether the user is at a traffic light intersection, whether the user is within a preset distance from a school or residential area, whether the user is at the head of a bridge, or whether the user is on a bridge. The user's current speed information includes speed in hours; The user profile information includes at least one of the following: short time, short distance, few navigation actions, or low cost.

10. The method according to claim 1, wherein, Before determining the route to be recommended, the process also includes: Get the current traffic information for the main route; When the road condition information of the current main route is determined to meet the conditions for replanning, the route to be recommended is determined.

11. A route navigation method, applied to a terminal device, comprising: Acquire and display the route to be recommended to be pushed to the terminal device, wherein the route to be recommended is determined by any one of the route recommendation methods described in claims 1-10; Receive recommendation acceptance data for the pushed routes to be recommended.

12. A computer program product for performing the route recommendation method of any one of claims 1-10 or the route navigation method of claim 11.

Images