Recommendation method, program product and recommendation system for interactive function of vehicle

Abstract

The invention relates to a recommendation method for interactive functions of a vehicle, and the recommendation method at least comprises the following steps: obtaining related data of each interactive function of the vehicle, the related data at least comprising interactive data of the interactive functions; obtaining a recommendation result about the interactive function by means of a recommendation model based on the related data; and highlighting the recommended interactive function on a user interface of the vehicle according to the recommendation result. The invention also relates to a corresponding computer program product and a recommendation system. The interactive function suitable for the vehicle user can be explored and recommended in a personalized manner, the utilization rate of the interactive function is remarkably improved, and the riding experience of the user is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle assisted driving, and in particular to a method for recommending interactive functions of a vehicle. The present invention also relates to a corresponding computer program product and a corresponding system for recommending interactive functions of a vehicle. Background Art

[0002] In recent years, with the advancement of technology and the improvement of living standards, people have higher and higher requirements for the intelligence and functionality of vehicles. To this end, more and more interactive functions are set in the vehicle's on-board system, such as automatic parking function, audio function, navigation function, air conditioning adjustment function, etc. These interactive functions allow vehicle users to interact with the on-board system, so as to perform desired operations in a simple way. The interactive functions can enhance the driving experience of vehicle users and increase the safety and convenience of the vehicle.

[0003] However, with the increase of interactive functions, some interactive functions are not fully used by users, and users are even unaware of the existence of these interactive functions, which significantly reduces the usage rate of the corresponding interchangeable functions and is inconsistent with their original development intention, and also affects the riding experience of vehicle users. Summary of the invention

[0004] Therefore, the purpose of the present invention is to propose an improved recommendation method for interactive functions of a vehicle, through which the interactive functions suitable for vehicle users can be discovered and recommended in a personalized manner, significantly improving the utilization rate of the interactive functions and improving the user's riding experience. The purpose of the present invention is also to propose a corresponding computer program product and a corresponding recommendation system for interactive functions of a vehicle.

[0005] According to a first aspect of the present invention, a method for recommending interactive functions of a vehicle is provided, wherein the recommendation method comprises at least the following steps:

[0006] S1: Acquire relevant data of each interactive function of the vehicle, wherein the relevant data at least includes interaction data of the interactive function;

[0007] S2: obtaining a recommendation result about the interactive function based on the relevant data with the help of a recommendation model;

[0008] S3: Highlighting the recommended interactive function on the user interface of the vehicle according to the recommendation result.

[0009] Compared with the prior art, in the recommendation method for interactive functions of a vehicle according to the present invention, relevant data of each interactive function of the vehicle are first obtained, and the relevant data can reflect the preferences of vehicle users for the interactive functions. Then, these relevant data are analyzed and processed with the help of a recommendation model to obtain personalized recommendation results about the interactive functions. The recommendation results may include interactive functions that are preferred by the user but have not been used or have a low usage rate, or interactive functions that are suitable for user use under specific conditions. Then, the interactive functions recommended in the recommendation results are highlighted on the user interface of the vehicle to remind the user to use these interactive functions, which can significantly improve the utilization rate of the interactive functions, simplify the user's operations and improve the user's riding experience.

[0010] Exemplarily, the interaction data includes the function ID, usage timestamp, usage duration, and usage frequency of the corresponding interactive function; and / or, the relevant data also includes at least one of the following data: driving data, positioning data, and environmental data of the vehicle.

[0011] Exemplarily, in step S3, the recommended interactive functions are highlighted in at least one of the following ways: increasing the brightness of the icons of the recommended interactive functions, decreasing the brightness of the icons of the non-recommended interactive functions, and listing the recommended interactive functions in the form of a pop-up window or a floating window.

[0012] Exemplarily, the recommendation model is a hybrid recommendation model formed by a collaborative filtering model and a content recommendation model, which is trained based on historical data of the relevant data.

[0013] Exemplarily, the implementation process of the recommendation model includes the following sub-steps:

[0014] S21: constructing a user function interaction matrix based on the relevant data;

[0015] S22: Implement collaborative filtering, and use matrix decomposition to decompose the user function interaction matrix into a user latent feature matrix and a function latent feature matrix to predict the user's preference for the interactive functions;

[0016] S23: Implement content recommendation, and recommend interactive functions similar to the preferred interactive functions based on feature similarity;

[0017] S24: Comprehensively analyzing the results of sub-steps S22 and S23, a recommendation result on the interactive function is obtained.

[0018] Exemplarily, in sub-step S22, matrix decomposition is performed by singular value decomposition or alternating least squares method; and / or, in sub-step S23, the feature similarity is calculated by the cosine similarity between the feature vectors of the interactive functions; and / or, in sub-step S24, weights are set for the results of sub-step S22 and sub-step S23, and the score of the interactive function is calculated by a linear combination of the results of these two sub-steps, thereby obtaining an optimized recommendation result.

[0019] Exemplarily, the recommendation method is implemented at a predetermined frequency; and / or, the recommendation method is implemented when the relevant data meets a predetermined condition.

[0020] Exemplarily, the recommendation method additionally includes step S4: the user of the vehicle inputs a confirmation command by touch or voice to start the interactive function recommended in step S3; and / or, the recommendation method additionally includes step S5: iteratively optimizing the recommendation model based on user feedback and behavior data.

[0021] According to a second aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by one or more processors, the processors are capable of performing the recommendation method according to the present invention.

[0022] According to a third aspect of the present invention, there is provided a recommendation system for interactive functions of a vehicle, wherein the recommendation system is configured to be suitable for implementing the recommendation method according to the present invention, wherein the recommendation system at least comprises:

[0023] - an acquisition module, the acquisition module being configured to acquire relevant data of each interactive function of the vehicle;

[0024] - a recommendation module, the recommendation module being configured to derive a recommendation result about the interactive function based on the relevant data with the help of a recommendation model; and

[0025] - a display module, wherein the display module is configured to highlight the recommended interactive functions according to the recommendation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention will be described in more detail below with reference to the accompanying drawings, so that the principles, features and advantages of the present invention can be better understood. The accompanying drawings include:

[0027] Figure 1 A schematic flow chart showing a method for recommending interactive functions of a vehicle according to an exemplary embodiment of the present invention;

[0028] Figure 2A schematic flow chart showing an implementation process of step S2 of the recommendation method according to an exemplary embodiment of the present invention;

[0029] Figure 3 A schematic block diagram of a recommendation system for interactive functions of a vehicle according to an exemplary embodiment of the present invention is shown. DETAILED DESCRIPTION

[0030] In order to make the technical problems, technical solutions and beneficial technical effects to be solved by the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the scope of protection of the present invention.

[0031] This specification provides method operation steps as described in the embodiments or flow charts, but more or fewer operation steps may be included based on routine or non-creative work. The order of steps listed in the embodiments is only one way of executing the steps among many, and does not represent the only execution order.

[0032] Figure 1 A schematic flow chart of a method for recommending interactive functions of a vehicle according to an exemplary embodiment of the present invention is shown.

[0033] like Figure 1 As shown, the recommended method according to the present invention comprises at least the following steps:

[0034] S1: Obtain relevant data of each interactive function of the vehicle, wherein the interactive function may include driving assistance functions, such as automatic parking, automatic driving or lane keeping, or multimedia and entertainment functions, such as audio functions, streaming media playback, or in-car adjustment functions, such as air conditioning or ambient light adjustment, etc. Of course, other interactive functions that other technicians in this field consider meaningful may also be considered, wherein the relevant data of the interactive function at least includes the interactive data of the interactive function, wherein the interactive data may include, for example, the function ID, usage timestamp, usage duration, and usage frequency of the corresponding interactive function, and the user's preference for the corresponding interactive function may be reflected through the interactive data. The longer the usage duration or the higher the usage frequency of a specific interactive function, the higher the user's preference for the interactive function, wherein the relevant data of the interactive function may be recorded and stored in the back-end server during the driving process of the vehicle;

[0035] S2: based on the acquired relevant data, a recommendation result about the interactive function is obtained by using a pre-trained recommendation model, wherein the relevant data is the input of the recommendation model, and the recommendation result including the recommended interactive function is the output of the recommendation model, and the recommended interactive function may be an interactive function that meets the user's preference, has not been used or has a low usage rate, or may be an interactive function that is suitable for the user to use under the current conditions;

[0036] S3: Highlighting the recommended interactive function on the user interface of the vehicle based on the recommendation result to prompt the user to use the recommended interactive function, wherein, for example, the interactive function can be highlighted by enlarging the icon of the recommended interactive function, which can make the user intuitively and easily notice the recommended interactive function; it is also possible to reduce the brightness of the icon of the non-recommended interactive function so that the icon of the non-recommended interactive function becomes darker or grayer, and accordingly, the icon of the recommended interactive function is highlighted; in addition, the recommended interactive function can also be listed in the form of a pop-up window or a floating window.

[0037] Therefore, by highlighting the recommended interactive functions on the user interface, users can more easily discover and use the interactive functions that meet their needs, which significantly increases the usage rate of the interactive functions and improves the user's riding experience.

[0038] Exemplarily, the relevant data of the interactive function includes at least one of the following data in addition to the interactive data: the driving data of the vehicle, the positioning data, and the environmental data. Here, the driving data may be the driving speed, gear, continuous driving time, charge level, etc. of the vehicle, and the environmental data may be the ambient temperature, weather, distance relative to surrounding obstacles, surrounding traffic conditions, etc. Of course, other data that other technicians in this field consider meaningful can also be considered. These data can reflect the conditions under which the user prefers to use a specific interactive function. For example, the automatic parking function of the vehicle can be associated with the driving speed, gear and positioning data of the vehicle, the air conditioning function of the vehicle can be associated with the ambient temperature, and the audio function of the vehicle can be associated with the continuous driving time of the vehicle and the degree of traffic congestion. Of course, other relevant data that other technicians in this field consider meaningful and their association with specific interactive functions can also be considered.

[0039] Figure 2A schematic flow chart of the implementation process of step S2 in the recommendation method according to an exemplary embodiment of the present invention is shown. Here, the recommendation model used in step S2 is especially a hybrid recommendation model formed by a collaborative filtering model and a content recommendation model, which is trained based on historical data of relevant data, wherein the collaborative filtering model can infer the user's interest in interactive functions that have not been used and provide personalized recommendations, while the content recommendation model can recommend new interactive functions that are similar to functions that the user has used, and the hybrid recommendation model formed thereby can provide accurate and diverse recommendation results.

[0040] like Figure 2 As shown, the implementation process of the recommendation model in step S2 includes the following sub-steps:

[0041] S21: constructing a user function interaction matrix based on the acquired relevant data, wherein the user function interaction matrix is ​​used to map the interaction relationship between the user and the interactive functions, so as to analyze the user's preference for different functions, wherein the preference may be expressed as the user's usage frequency or usage duration for the corresponding interactive functions;

[0042] S22: Implement collaborative filtering, wherein the user-function interaction matrix is ​​decomposed into a user latent feature matrix and a function latent feature matrix by using matrix decomposition, so as to predict the user's preference for the interactive functions, wherein the matrix decomposition can be performed, for example, in the form of singular value decomposition (SVD) or alternating least squares (ALS). Of course, other matrix decomposition methods that are considered meaningful by those skilled in the art may also be considered, such as non-negative matrix decomposition and probabilistic matrix decomposition;

[0043] S23: implementing content recommendation, recommending interactive functions similar to the preferred interactive functions based on feature similarity, wherein the attributes of each interactive function are converted or extracted into a machine-readable feature vector, and the feature similarity is calculated by the cosine similarity between the feature vectors of the interactive functions;

[0044] S24: The results of sub-steps S22 and S23 are comprehensively analyzed to obtain recommendation results about the interactive functions, which can fully utilize the advantages of the collaborative filtering model and the content recommendation model and make up for the shortcomings of a single model to obtain accurate and diverse recommendation results, wherein weights are particularly set for the results of sub-steps S22 and S23, and the final score of each interactive function is calculated by a linear combination or weighted summation of the results of these two sub-steps. The higher the score, the higher the recommendation degree of the corresponding interactive function, thereby obtaining an optimized recommendation result.

[0045] Exemplarily, the recommendation method according to the present invention is implemented at a predetermined frequency, such as weekly or monthly, to prompt the user to use new or underutilized interactive functions at the frequency. In addition, it can also be considered that the recommendation method according to the present invention is implemented when the relevant data of the interactive function meets the predetermined conditions. Specifically, for example, when the ambient temperature of the vehicle reaches the temperature threshold, the user is recommended to use the air conditioning adjustment function or a similar interactive function through the implementation of the recommendation method, or, when the positioning data of the vehicle is in the parking lot and the gear position of the vehicle is in the reverse gear, the user is recommended to use the automatic parking function through the implementation of the recommendation method.

[0046] For example, Figure 1 As shown, the recommendation method additionally includes step S4: the user of the vehicle inputs a confirmation instruction by touching the display screen to start the interactive function recommended in step S3. It is also possible that the user inputs the confirmation instruction by voice, such as "confirm to start".

[0047] For example, Figure 1 As shown, the recommendation method additionally includes step S5: iteratively optimizing the recommendation model according to user feedback and behavior data, wherein the feedback can be collected, for example, in the form of a questionnaire or interview, wherein the effectiveness of the recommendation result can be evaluated by comparing the user behavior data before and after the implementation of the recommendation method. Through iterative optimization of the recommendation model, interactive functions can be recommended more targetedly and the user's riding experience can be continuously improved.

[0048] Figure 3 A schematic block diagram of a recommendation system 100 for interactive functions of a vehicle according to an exemplary embodiment of the present invention is shown. Here, the recommendation system 100 can implement the recommendation method according to the present invention by means of a computer program product according to the present invention, wherein the computer program product includes a computer program, and when the computer program is executed by one or more processors, the processor can execute the recommendation method according to the present invention.

[0049] like Figure 3 As shown, the recommendation system 100 includes an acquisition module 10, which is configured to acquire relevant data of each interactive function of the vehicle, wherein the acquisition module 10 is particularly provided with an API interface for each interactive function to acquire and record the interactive data of the corresponding interactive function, such as the use timestamp, function operation, etc. In addition, the acquisition module 10 can also acquire other relevant data, such as the driving data, positioning data, environmental data, etc. of the vehicle.

[0050] like Figure 3As shown, the recommendation system 100 includes a recommendation module 20, which receives relevant data from the acquisition module 10 and is configured to derive recommendation results about interactive functions based on the relevant data with the help of a recommendation model, wherein the recommendation model can be pre-trained and stored in the recommendation module 20.

[0051] like Figure 3 As shown, the recommendation system 100 includes a display module 30, which is configured to receive the recommendation results from the recommendation module 20 and highlight the recommended interactive functions according to the recommendation results. Here, the display module 30 may include a dashboard, a central control display screen, or a passenger display screen.

[0052] The above explanation of the embodiments only describes the invention within the framework of the examples. Of course, the individual features of the embodiments can be freely combined with one another as long as it makes technical sense, without departing from the framework of the invention.

[0053] Other advantages and alternative embodiments of the present invention are obvious to those skilled in the art. Therefore, the present invention in its broader sense is not limited to the specific details, representative structures and exemplary embodiments shown and described. On the contrary, those skilled in the art can make various modifications and substitutions without departing from the basic spirit and scope of the present invention.

Claims

1. A method for recommending interactive functions of a vehicle, characterized in that: The recommended method comprises at least the following steps: S1: Acquire relevant data of each interactive function of the vehicle, wherein the relevant data at least includes interaction data of the interactive function; S2: obtaining a recommendation result about the interactive function based on the relevant data with the help of a recommendation model; S3: Highlighting the recommended interactive function on the user interface of the vehicle according to the recommendation result.

2. The recommendation method according to claim 1, characterized in that: The interaction data includes the function ID, usage timestamp, usage duration, and usage frequency of the corresponding interactive function; and / or The relevant data also includes at least one of the following data: driving data, positioning data, and environmental data of the vehicle.

3. The recommendation method according to claim 1 or 2, characterized in that: In step S3, the recommended interactive functions are highlighted in at least one of the following ways: increasing the brightness of the icons of the recommended interactive functions, decreasing the brightness of the icons of the non-recommended interactive functions, and listing the recommended interactive functions in the form of a pop-up window or a floating window.

4. The recommendation method according to any one of the preceding claims, characterized in that The recommendation model is a hybrid recommendation model trained based on historical data of the relevant data and formed by a collaborative filtering model and a content recommendation model.

5. The recommendation method according to claim 4, characterized in that: The implementation process of the recommendation model includes the following sub-steps: S21: constructing a user function interaction matrix based on the relevant data; S22: Implement collaborative filtering, and use matrix decomposition to decompose the user function interaction matrix into a user latent feature matrix and a function latent feature matrix to predict the user's preference for the interactive functions; S23: Implement content recommendation, and recommend interactive functions similar to the preferred interactive functions based on feature similarity; S24: Comprehensively analyzing the results of sub-steps S22 and S23, a recommendation result on the interactive function is obtained.

6. The recommendation method according to claim 5, characterized in that: In sub-step S22, matrix decomposition is performed by singular value decomposition or alternating least squares method; and / or In sub-step S23, the feature similarity is calculated by the cosine similarity between the feature vectors of the interactive functions; and / or In sub-step S24, weights are set for the results of sub-step S22 and sub-step S23, and the scores of the interactive functions are calculated through a linear combination of the results of these two sub-steps, thereby obtaining an optimized recommendation result.

7. Recommendation method according to any one of the preceding claims, characterized in that The recommended method is implemented at a predetermined frequency; and / or When the relevant data meets a predetermined condition, the recommendation method is implemented.

8. Recommendation method according to any one of the preceding claims, characterized in that The recommendation method additionally comprises step S4: the user of the vehicle inputs a confirmation instruction by touch or voice to start the interactive function recommended in step S3; and / or The recommendation method additionally comprises step S5: iteratively optimizing the recommendation model according to user feedback and behavior data.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by one or more processors, the processors are capable of performing the recommendation method according to any one of claims 1 to 8.

10. A recommendation system (100) for interactive functions of a vehicle, characterized in that: The recommendation system (100) is configured to be suitable for implementing the recommendation method according to any one of claims 1 to 8, wherein the recommendation system (100) at least comprises: - an acquisition module (10), which is configured to acquire relevant data of each interactive function of the vehicle; - a recommendation module (20), the recommendation module being configured to derive a recommendation result about the interactive function based on the relevant data with the help of a recommendation model; and - a display module (30), the display module being configured to highlight the recommended interactive functions according to the recommendation results.

Images