Vehicle machine application preference analysis method and device, vehicle and storage medium

Abstract

The invention relates to the technical field of data processing, in particular to a vehicle-mounted terminal application preference analysis method and device, a vehicle and a storage medium. The method comprises the steps of performing target data processing on driving state data of a vehicle and at least one of use frequency data, use duration data, switching frequency data and application score data of a user for each vehicle machine application in the vehicle to obtain processed multi-dimensional data; and inputting the processed multi-dimensional data into a pre-constructed target deep learning model, analyzing the use preference of each vehicle-mounted information entertainment product application to generate a use preference value of each vehicle-mounted information entertainment product application, and determining a vehicle-mounted information entertainment product application preference analysis result of the vehicle according to the use preference value. Therefore, the problems that in the prior art, the dimension of collected user data is relatively single, deep understanding of user preferences is lacked, complex driving scenes cannot be dynamically adapted, the accuracy and personalized service ability of user preference analysis are reduced, and personalized requirements of users cannot be met are solved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, vehicle and storage medium for analyzing vehicle application preferences. Background Art

[0002] With the rapid development of smart cars and Internet of Vehicles technology, in-vehicle infotainment systems, navigation systems and driver assistance systems have gradually become richer, and user needs are becoming increasingly diverse. However, different users have significant differences in their needs and preferences for car applications. How to intelligently analyze and recommend appropriate car applications based on user habits and behavior data has become a key issue in improving user experience and the competitiveness of automakers.

[0003] In the related technology, user data is obtained from the vehicle terminal, that is, the original recorded data representing the user's daily life is collected, and then the user data is cleaned to filter out invalid data to obtain filtered structured valid user data, so as to perform cluster analysis on the valid user data to obtain user preference characteristics.

[0004] However, the dimensions of user data collected in related technologies are relatively single, lacking a deep understanding of user preferences and unable to dynamically adapt to complex driving scenarios. This reduces the accuracy of user preference analysis and personalized service capabilities, and is unable to meet users' personalized needs, which needs to be urgently addressed. Summary of the invention

[0005] The present application provides a method, device, vehicle and storage medium for analyzing vehicle application preferences to solve the problem that the dimensions of user data collected in related technologies are relatively single, there is a lack of in-depth understanding of user preferences, and it is impossible to dynamically adapt to complex driving scenarios, which reduces the accuracy of user preference analysis and personalized service capabilities, and cannot meet the personalized needs of users.

[0006] The first aspect of the present application provides a method for analyzing vehicle application preferences, comprising the following steps: collecting vehicle driving status data and user usage frequency data, usage duration data, switching frequency data and application rating data for each vehicle application in the vehicle; performing target data processing on the driving status data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application rating data to obtain processed multidimensional data; inputting the processed multidimensional data into a pre-built target deep learning model, analyzing the usage preference of each vehicle application to generate a usage preference value for each vehicle application, and determining the vehicle application preference analysis result of the vehicle based on the usage preference value.

[0007] Optionally, in one embodiment of the present application, the target data processing of the driving status data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application rating data to obtain processed multidimensional data includes: target denoising processing of at least one of the position information, driving speed and road condition information in the driving status data to obtain first processed data; target data cleaning of at least one of the usage frequency data, the usage duration data, the switching frequency data and the application rating data to obtain second processed data; and determining the processed multidimensional data using the first processed data and the second processed data.

[0008] Optionally, in one embodiment of the present application, the analyzing the usage preference of each vehicle-mounted application to generate the usage preference value of each vehicle-mounted application includes: based on the pre-built target deep learning model, analyzing the processed multidimensional data to generate a usage preference vector for each vehicle-mounted application; and determining the usage preference value of each vehicle-mounted application using the usage preference vector.

[0009] Optionally, in one embodiment of the present application, the use of the usage preference vector to determine the usage preference value of each vehicle-mounted application includes: determining the recommendation score of each vehicle-mounted application based on the target application feature vector in the vehicle's Internet of Vehicles ecosystem and the usage preference vector; and determining the usage preference value of each vehicle-mounted application based on the recommendation score of each vehicle-mounted application.

[0010] Optionally, in one embodiment of the present application, after determining the vehicle's vehicle application preference analysis result according to the usage preference value, it also includes: generating a recommended vehicle application layout for the vehicle according to the vehicle application preference analysis result; sending the recommended vehicle application layout to a preset terminal to display the recommended vehicle application layout on the preset terminal.

[0011] The second aspect of the present application provides a device for analyzing vehicle application preferences, including: a collection module, used to collect vehicle driving status data and user usage frequency data, usage duration data, switching frequency data and application score data for each vehicle application in the vehicle; an acquisition module, used to perform target data processing on the driving status data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain processed multidimensional data; an analysis module, used to input the processed multidimensional data into a pre-built target deep learning model, analyze the usage preference of each vehicle application to generate a usage preference value for each vehicle application, and determine the vehicle application preference analysis result of the vehicle according to the usage preference value.

[0012] Optionally, in one embodiment of the present application, the acquisition module includes: a first processing unit, used to perform target denoising processing on at least one of the position information, driving speed and road condition information in the driving status data to obtain first processed data; a second processing unit, used to perform target data cleaning on at least one of the usage frequency data, the usage time data, the switching frequency data and the application rating data to obtain second processed data; a first determination unit, used to determine the processed multidimensional data using the first processed data and the second processed data.

[0013] Optionally, in one embodiment of the present application, the analysis module includes: an analysis unit, used to analyze the processed multidimensional data based on the pre-built target deep learning model to generate a usage preference vector for each vehicle-mounted application; and a second determination unit, used to determine the usage preference value of each vehicle-mounted application using the usage preference vector.

[0014] Optionally, in one embodiment of the present application, the analysis module includes: a third determination unit, used to determine the recommendation score of each vehicle application based on the target application feature vector and the usage preference vector in the vehicle's Internet of Vehicles ecosystem; a fourth determination unit, used to determine the usage preference value of each vehicle application according to the recommendation score of each vehicle application.

[0015] Optionally, in one embodiment of the present application, the device of the embodiment of the present application also includes: a generation module, which is used to generate a recommended layout of vehicle-machine applications for the vehicle according to the vehicle-machine application preference analysis result after determining the vehicle-machine application preference analysis result of the vehicle according to the usage preference value; and a sending module, which is used to send the recommended layout of vehicle-machine applications to a preset terminal after determining the vehicle-machine application preference analysis result of the vehicle according to the usage preference value, so as to display the recommended layout of vehicle-machine applications on the preset terminal.

[0016] A third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for analyzing vehicle application preferences as described in the above embodiment.

[0017] The fourth aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above-mentioned method for analyzing vehicle application preferences.

[0018] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for analyzing vehicle application preferences.

[0019] The embodiment of the present application can process the collected driving status data of the vehicle with at least one of the frequency data, usage duration data, switching frequency data and application rating data of each vehicle-mounted application in the vehicle by the user to obtain processed multi-dimensional data, which is then input into a pre-built target deep learning model to analyze the usage preference of each vehicle-mounted application to generate a usage preference value for each vehicle-mounted application, and determine the vehicle-mounted application preference analysis result of the vehicle based on the usage preference value, effectively improving the accuracy of user preference analysis and personalized service capabilities. As a result, the problems of the relatively single dimension of user data collection in the related technology, the lack of in-depth understanding of user preferences, the inability to dynamically adapt to complex driving scenarios, and the reduced accuracy of user preference analysis and personalized service capabilities are solved.

[0020] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0022] Figure 1 A structural diagram of a system for analyzing vehicle machine application preferences provided according to an embodiment of the present application;

[0023] Figure 2 A flowchart of a method for analyzing vehicle machine application preferences provided according to an embodiment of the present application;

[0024] Figure 3 A schematic diagram of the structure of a device for analyzing vehicle machine application preferences provided according to an embodiment of the present application;

[0025] Figure 4 It is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0027] The following describes the method, device, vehicle and storage medium for analyzing the preferences of vehicle-mounted applications in the embodiment of the present application with reference to the accompanying drawings. In view of the problem that the dimensions of collecting user data in the related technologies mentioned in the above background technology are relatively single, lack a deep understanding of user preferences, cannot dynamically adapt to complex driving scenarios, reduce the accuracy of user preference analysis and personalized service capabilities, and cannot meet the personalized needs of users, the present application provides a method for analyzing the preferences of vehicle-mounted applications, in which the collected driving state data of the vehicle and at least one of the frequency data, usage time data, switching frequency data and application score data of each vehicle-mounted application in the vehicle can be processed by target data to obtain the processed multidimensional data, which is then input into the pre-built target deep learning model to analyze the usage preferences of each vehicle-mounted application to generate the usage preference value of each vehicle-mounted application, and determine the vehicle-mounted application preference analysis result of the vehicle according to the usage preference value, effectively improving the accuracy of the user preference analysis and personalized service capabilities. Therefore, the problem that the dimensions of collecting user data in the related technologies are relatively single, lack a deep understanding of user preferences, cannot dynamically adapt to complex driving scenarios, reduce the accuracy of the user preference analysis and personalized service capabilities, etc. are solved.

[0028] like Figure 1 As shown, an embodiment of the present application establishes a system for analyzing vehicle application preference, which includes a data acquisition module, a big data cloud platform, a data processing and analysis module, a deep learning modeling module and a preference analysis and recommendation module.

[0029] Among them, the data collection module: collects user application usage behavior, vehicle status data (such as location, speed, road conditions), network environment and other information in real time through the on-board 4G / 5G communication module, T-BOX and central control screen.

[0030] Big data cloud platform: The collected data is transmitted to the Internet of Vehicles big data cloud platform for storage and processing. The platform supports large-scale data storage, distributed computing and real-time analysis.

[0031] Data processing and analysis module: performs data cleaning and denoising on the collected data.

[0032] Deep learning modeling module: Utilize the computing power of the big data platform to build a Transformer-based deep learning model, and combine user usage habits, vehicle scenario data, and historical preferences to perform multimodal data fusion and preference modeling.

[0033] Preference analysis and recommendation module: Calculate the user's preference values ​​for various types of vehicle-mounted applications based on the modeling results, and generate a recommendation report for reference by users or OEMs.

[0034] Specifically, Figure 2 A flowchart of a method for analyzing vehicle computer application preferences provided in an embodiment of the present application.

[0035] like Figure 2 As shown, the method for analyzing vehicle computer application preference includes the following steps:

[0036] In step S201, the driving status data of the vehicle and the usage frequency data, usage duration data, switching frequency data and application rating data of each vehicle application in the vehicle by the user are collected.

[0037] It can be understood that the embodiments of the present application can collect the vehicle's driving status data and the user's usage frequency data, usage duration data, switching frequency data and application rating data for each car-machine application in the vehicle. For example, the vehicle's position, speed, current driving conditions, etc. can be collected through the data collection module in the above-mentioned system. The user's application usage data under different driving conditions can also be collected, including usage frequency, usage duration, switching frequency and application rating, etc. For example, when the user is driving on an elevated road in the city, he is accustomed to using an audiobook application, etc., which effectively improves the feasibility of the car-machine preference analysis.

[0038] In step S202, target data processing is performed on the driving state data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain processed multidimensional data.

[0039] It can be understood that the embodiment of the present application can perform target data processing on the collected driving status data and at least one of the usage frequency data, usage duration data, switching frequency data and application score data through the data processing and analysis module in the above system, such as cleaning and denoising the data to obtain multi-dimensional data with a more standard format, wherein the data format table of user application usage is shown in Table 1, and the specific Table 1 is as follows:

[0040] Table 1

[0041]

[0042] Among them, Table 2 is a table of user core parameters. The judgment of user preferences is mainly based on the core parameters in Table 2. The specific Table 2 is as follows:

[0043] Table 2

[0044]

[0045]

[0046] Among them, in one embodiment of the present application, target data processing is performed on the driving status data and at least one of the usage frequency data, usage duration data, switching frequency data and application score data to obtain processed multidimensional data, including: target denoising processing is performed on at least one of the position information, driving speed and road condition information in the driving status data to obtain first processed data; target data cleaning is performed on at least one of the usage frequency data, usage duration data, switching frequency data and application score data to obtain second processed data; and the processed multidimensional data is determined using the first processed data and the second processed data.

[0047] During the actual implementation process, the embodiment of the present application can denoise at least one of the position information, driving speed and road condition information in the driving status data through the data processing and analysis module in the above-mentioned system to obtain first processed data. Then, at least one of the usage frequency data, usage time data, switching frequency data and application score data can be cleaned to obtain second processed data, so as to determine the processed multidimensional data, wherein the specific type of cleaning or denoising should be determined based on the collected data, which helps to improve the quality of the data and promote more accurate and reliable analysis results.

[0048] In step S203, the processed multidimensional data is input into a pre-built target deep learning model, and the usage preference of each vehicle-mounted application is analyzed to generate a usage preference value for each vehicle-mounted application, and the vehicle-mounted application preference analysis result of the vehicle is determined based on the usage preference value.

[0049] It can be understood that the embodiments of the present application can input the processed multi-dimensional data into the target deep learning model pre-constructed in the following steps, analyze the usage preferences and operating behaviors of each car-machine application, etc., to generate a usage preference value for each car-machine application, and determine the vehicle-machine application preference analysis results of the vehicle based on the usage preference value, and determine the user's preference level for each car-machine application, so that the car-machine applications can be sorted according to the user's preference level, effectively improving the user's personalized experience.

[0050] In the embodiment of the present application, the target deep learning model can be a Transformer-based deep learning model, that is, the above system can use the Transformer architecture to model user behavior and preferences, and the specific design is as follows:

[0051] (1) Input embedding layer: Encode user behavior data (such as application usage sequence), vehicle status data, and scene data. First, each data point is converted into an embedding vector (embedding), and positional encoding is introduced so that the Transformer can understand sequential information.

[0052] (2) Positional encoding: Although Transformer abandons the cyclic structure of processing sequence data in traditional recurrent neural networks, the order of sequence data is still crucial. To this end, Transformer adds positional encoding to the input embedding, encoding each position through fixed sine and cosine functions, so that the model can capture the order information of the input data.

[0053] (3) Multi-head attention mechanism: The self-attention mechanism is the core of Transformer. It allows the model to dynamically pay attention to other related elements in the sequence and calculate their importance when processing the current element. The multi-head self-attention mechanism captures different dependencies and features through multiple parallel attention heads. Each attention head calculates the attention weight in a different subspace and then merges the results. The attention weight between each input element is calculated by the following formula. The multi-head mechanism further linearly transforms the results and then splices them, so that features at different levels can be captured.

[0054]

[0055] Among them, Q represents the query vector, K represents the key vector, V represents the value vector, and d k Indicates the dimension of the key vector.

[0056] (4) Feedforward network: Each encoder and decoder layer also contains a fully connected FFN (Feed-Forward Network), which processes the vector at each position independently. FFN usually contains two linear transformations, with a ReLU activation function in between.

[0057] (5) Residual connection and layer normalization: To prevent the gradient from vanishing or exploding in deep networks, Transformer adds a residual connection to each sublayer (self-attention and feedforward network), that is, the input directly skips the sublayer and is added to the output. At the same time, layer normalization is used to stabilize the model training process.

[0058] (6) Encoder-decoder interaction: The encoder of Transformer is composed of multiple identical encoder layers stacked together, each of which contains the self-attention mechanism and feedforward network mentioned above. The encoder processes the input sequence and extracts global features. When generating the output sequence, the decoder not only processes the decoding results of the previous steps, but also combines the output of the encoder and interacts through the "encoder-decoder attention mechanism" to generate the final prediction result.

[0059] In one embodiment of the present application, the usage preference of each vehicle-mounted application is analyzed to generate a usage preference value for each vehicle-mounted application, including: based on a pre-built target deep learning model, analyzing the processed multidimensional data to generate a usage preference vector for each vehicle-mounted application; and using the usage preference vector to determine the usage preference value of each vehicle-mounted application.

[0060] For example, the embodiment of the present application can analyze the user's operation behavior and application preferences based on the pre-built target deep learning model, as follows:

[0061] (1) Input data embedding: The user’s operational behavior data (such as downloading, deleting, moving, star rating, etc.) is converted into a high-dimensional embedding vector, which retains the semantic information of each behavior.

[0062] (2) Self-attention mechanism to model behavioral dependencies: Through the multi-head self-attention mechanism, the model can identify the potential relationship between different operations in user behavior data. For example, there may be a high correlation between the frequent changes in the application location and the usage time, and this correlation is captured by the attention mechanism.

[0063] (3) Feature extraction and preference vector generation: The encoder layer extracts features from the input sequence and generates user preference vectors. These preference vectors will be used to calculate the user's interest score for each application and determine the priority in the recommendation list.

[0064] (4) Recommendation output and feedback update: Based on the preference scores, the above system generates an application recommendation list and uses the user's actual usage as feedback to continuously optimize the parameters of the Transformer model.

[0065] Optionally, in one embodiment of the present application, a usage preference vector is used to determine a usage preference value for each vehicle-mounted application, including: determining a recommendation score for each vehicle-mounted application based on a target application feature vector and a usage preference vector in the vehicle's Internet of Vehicles ecosystem; and determining a usage preference value for each vehicle-mounted application based on the recommendation score of each vehicle-mounted application.

[0066] As a possible implementation method, in the Internet of Vehicles ecosystem, in order to ensure the best experience of users in car applications, this application generates a user preference vector by comprehensively analyzing multi-dimensional data such as user operation data, application usage habits, and application star ratings. Based on the user preference vector generated by the Transformer model, the system further scores various applications in the ecosystem and finally calculates the user's preference value. The following is a detailed discussion of the specific preference analysis and application recommendation process:

[0067] Among them, the generation and calculation of preference vectors: The core of preference analysis is to generate a high-dimensional vector that represents user preferences. Through the multi-head self-attention mechanism in the Transformer model, the system can capture data associations in many aspects such as user operation behavior, usage habits, and application star ratings, thereby deriving the user's preference for various applications. Preference vector P u It represents the preference characteristics of user u for various applications, which is generated by the following data:

[0068] (1) Operation behavior embedding: Various operations performed by users on the application on the car screen, such as downloading, deleting, updating, and adjusting the position, are encoded into an embedding vector E o .

[0069] (2) Application usage habits embedding: Application usage time, switching frequency, star rating and other data are embedded into vector E h .

[0070] (3) Position and layout information embedding: Generate a position feature vector E according to the layout positions of different levels (negative first screen, first screen, second screen, etc.) on the large screen of the vehicle computer. l .

[0071] These embedding vectors are input into the Transformer model, and after being processed by multiple layers of encoders, a comprehensive preference vector P is output. u , its formula is expressed as:

[0072] P u = Transformer(E o ,E h ,E l )

[0073] Among them, E o represents the embedding vector, E h represents the embedded vector, E l represents the position feature vector, P u represents the preference vector.

[0074] The preference vector contains the user's preference feature information for different types of applications and can be dynamically updated to adapt to changes in user preferences.

[0075] Next, based on the generated preference vector P u , the system converts the application feature vector A in the Internet of Vehicles ecosystem i Match it with the user preference vector and calculate the user's recommendation score S for each application i :

[0076] S i =P u ·A i +b i

[0077] Among them, A i represents the feature vector of application i, b i is the bias term of the model, used to adjust the baseline score of the application. i The higher the value, the stronger the user's preference for the app and the higher the recommendation priority.

[0078] In the specific implementation, the system combines the following strategies to make application recommendations:

[0079] (1) Prioritization based on ratings: The system ranks the recommendation scores of all apps and prioritizes apps with the highest scores. For example, an app with a score of 0.9 will be ranked first in the recommendation list.

[0080] (2) Diversity of application types: In order to meet the various needs of users in the car environment, the system will balance the types of applications in the recommendation results. For example, it provides balanced recommendations among various types of applications such as navigation, entertainment, and social networking.

[0081] (3) Personalized layout optimization: Based on user preferences and screen layout habits, the system provides personalized application layout suggestions. For example, frequently used applications are automatically adjusted to the negative one screen or the first screen of the car screen to improve user convenience.

[0082] Based on the rating results, the system calculates the user's preference value U for a certain application. i :

[0083] U i =σ(S i )

[0084] Among them, σ is a normalization function (such as Sigmoid function or Softmax function), which converts the recommendation score S iConverted into a preference value between 0 and 1. The closer the preference value is to 1, the higher the user's preference for the application. The final preference value can be divided into the following levels. Table 3 is a level table of user preference corresponding values, as shown in Table 3:

[0085] Table 3

[0086] High Preference The preference value is above 0.8. Medium Preference The preference value is between 0.5 and 0.8. Low Preference Preference values ​​are below 0.5.

[0087] As shown in Table 4, it is an example table of output results, which shows some of the results output by the system after preference analysis and application recommendation. The specific Table 4 is as follows:

[0088] Table 4

[0089] Application ID (AppID) Recommendation score(Score) User preference Navigation 0.92 high Music 0.85 high News 0.78 middle Social Media 0.60 middle Weather 0.65 middle

[0090] As shown in Table 4, the scores of Navigation and Music applications are relatively high, indicating that users have a strong preference for these applications, while the preference for Social Media News and Weather applications is relatively low.

[0091] Secondly, in order to continuously optimize the recommendation results, this application also designs a dynamic feedback mechanism:

[0092] 1. User feedback collection: Through pop-up windows or questionnaires, guide users to evaluate the recommendation results. User feedback data will be used to further train and optimize the Transformer model.

[0093] 2. Model adaptive update: As user preferences and operating habits change, the system will regularly retrain the model to ensure that the recommendation results are consistent with the user's current needs.

[0094] 3. Application feature library update: With the continuous update and iteration of applications in the Internet of Vehicles ecosystem, the system will dynamically update the application feature library to ensure that the recommendation results cover the latest applications that best meet user needs. Among them, Table 5 is an evaluation index table, and the specific Table 5 is as follows:

[0095] Table 5

[0096] Evaluation indicators Numeric Precision 88％ Recall 82％ Customer satisfaction 85％ Response time (ms) 350 Model update frequency Weekly Updates

[0097] Optionally, in one embodiment of the present application, after determining the vehicle's vehicle-machine application preference analysis result based on the usage preference value, it also includes: generating the vehicle's vehicle-machine application recommended layout based on the vehicle-machine application preference analysis result; sending the vehicle-machine application recommended layout to a preset terminal to display the vehicle-machine application recommended layout on the preset terminal.

[0098] For example, an embodiment of the present application can generate a recommended layout of vehicle computer applications based on the results of the vehicle computer application preference analysis, and send the recommended layout of vehicle computer applications to the user's vehicle side and the OEM's computer side, so that the user can use the recommended personalized vehicle computer application layout. At the same time, the OEM can also adjust the priority of vehicle computer applications in a timely manner according to the user's preferences, reduce development costs, improve the usability and efficiency of the vehicle computer system, and provide users with more considerate services to enhance the overall driving experience.

[0099] According to the method of analyzing vehicle application preferences proposed in the embodiment of the present application, the collected vehicle driving status data and at least one of the frequency data, usage duration data, switching frequency data, and application rating data of each vehicle application in the vehicle can be subjected to target data processing to obtain processed multi-dimensional data, which is then input into a pre-built target deep learning model to analyze the usage preferences of each vehicle application to generate a usage preference value for each vehicle application, and determine the vehicle application preference analysis result of the vehicle based on the usage preference value, effectively improving the accuracy of user preference analysis and personalized service capabilities. As a result, the problems of the relatively single dimension of collecting user data in the related technology, the lack of in-depth understanding of user preferences, the inability to dynamically adapt to complex driving scenarios, and the reduced accuracy of user preference analysis and personalized service capabilities are solved.

[0100] Next, a device for analyzing vehicle application preferences according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0101] Figure 3 It is a block diagram of a device for analyzing vehicle application preference according to an embodiment of the present application.

[0102] like Figure 3 As shown, the vehicle-mounted computer application preference analysis device 10 includes: a collection module 100 , an acquisition module 200 and an analysis module 300 .

[0103] Specifically, the collection module 100 is used to collect the driving status data of the vehicle and the user's usage frequency data, usage duration data, switching frequency data and application rating data of each vehicle application in the vehicle.

[0104] The acquisition module 200 is used to perform target data processing on the driving status data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain processed multi-dimensional data.

[0105] The analysis module 300 is used to input the processed multidimensional data into a pre-built target deep learning model, analyze the usage preference of each vehicle-mounted application to generate a usage preference value for each vehicle-mounted application, and determine the vehicle-mounted application preference analysis result of the vehicle according to the usage preference value.

[0106] Optionally, in one embodiment of the present application, the acquisition module 200 includes: a first processing unit, a second processing unit and a first determination unit.

[0107] The first processing unit is used to perform target denoising processing on at least one of the position information, the driving speed and the road condition information in the driving state data to obtain first processed data.

[0108] The second processing unit is used to perform target data cleaning on at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain second processed data.

[0109] The first determining unit is used to determine the processed multi-dimensional data by using the first processed data and the second processed data.

[0110] Optionally, in one embodiment of the present application, the analysis module 300 includes: an analysis unit and a second determination unit.

[0111] Among them, the analysis unit is used to analyze the processed multi-dimensional data based on a pre-built target deep learning model to generate a usage preference vector for each vehicle application.

[0112] The second determining unit is used to determine the usage preference value of each vehicle-mounted application by using the usage preference vector.

[0113] Optionally, in one embodiment of the present application, the analysis module 300 includes: a third determination unit and a fourth determination unit.

[0114] Among them, the third determination unit is used to determine the recommendation score of each vehicle application based on the target application feature vector and usage preference vector in the vehicle's Internet of Vehicles ecosystem.

[0115] The fourth determining unit is used to determine the usage preference value of each vehicle-mounted application according to the recommendation score of each vehicle-mounted application.

[0116] Optionally, in one embodiment of the present application, the device 10 of the embodiment of the present application further includes: a generating module and a sending module.

[0117] Among them, the generation module is used to generate a recommended layout of vehicle-mounted applications of the vehicle according to the vehicle-mounted application preference analysis result after determining the vehicle-mounted application preference analysis result according to the usage preference value.

[0118] The sending module is used to send the recommended layout of the vehicle-machine application to a preset terminal after determining the vehicle-machine application preference analysis result of the vehicle according to the usage preference value, so as to display the recommended layout of the vehicle-machine application on the preset terminal.

[0119] It should be noted that the above explanation of the embodiment of the method for analyzing the preference of vehicle-mounted computer applications is also applicable to the device for analyzing the preference of vehicle-mounted computer applications in this embodiment, and will not be repeated here.

[0120] According to the device for analyzing the preferences of vehicle-mounted computer applications proposed in the embodiment of the present application, the collected driving status data of the vehicle and at least one of the frequency data, usage duration data, switching frequency data, and application rating data of each vehicle-mounted computer application in the vehicle can be subjected to target data processing to obtain processed multi-dimensional data, which is then input into a pre-built target deep learning model to analyze the usage preferences of each vehicle-mounted computer application to generate a usage preference value for each vehicle-mounted computer application, and determine the vehicle-mounted computer application preference analysis result of the vehicle based on the usage preference value, thereby effectively improving the accuracy of the user preference analysis and the personalized service capability. Thus, the problems of the relatively single dimension of collecting user data in the related technology, the lack of in-depth understanding of user preferences, the inability to dynamically adapt to complex driving scenarios, and the reduced accuracy of the user preference analysis and the personalized service capability are solved.

[0121] Figure 4 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0122] Memory 401 , processor 402 , and a computer program stored in the memory 401 and executable on the processor 402 .

[0123] When the processor 402 executes the program, the method for analyzing the vehicle machine application preference provided in the above embodiment is implemented.

[0124] Furthermore, the vehicle also includes:

[0125] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0126] The memory 401 is used to store computer programs that can be executed on the processor 402 .

[0127] The memory 401 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0128] If the memory 401, the processor 402 and the communication interface 403 are implemented independently, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0129] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0130] The processor 402 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0131] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for analyzing vehicle application preferences as described above is implemented.

[0132] This embodiment also provides a computer program product, including a computer program, which, when executed, is used to implement the above method for analyzing vehicle application preferences.

[0133] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0134] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0135] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0136] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0137] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0138] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0139] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0140] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for analyzing vehicle computer application preference, characterized in that: The following steps are involved: Collecting vehicle driving status data and user usage frequency data, usage duration data, switching frequency data and application rating data for each vehicle computer application in the vehicle; Performing target data processing on the driving state data and at least one of the usage frequency data, the usage duration data, the switching frequency data, and the application score data to obtain processed multidimensional data; The processed multi-dimensional data is input into a pre-built target deep learning model, the usage preference of each vehicle-mounted application is analyzed to generate a usage preference value for each vehicle-mounted application, and the vehicle-mounted application preference analysis result of the vehicle is determined according to the usage preference value.

2. The method according to claim 1, characterized in that The step of performing target data processing on the driving state data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain processed multi-dimensional data includes: Performing target denoising processing on at least one of the position information, the driving speed and the road condition information in the driving state data to obtain first processed data; performing target data cleaning on at least one of the usage frequency data, the usage duration data, the switching frequency data, and the application rating data to obtain second processed data; The processed multi-dimensional data is determined using the first processed data and the second processed data.

3. The method according to claim 1, characterized in that The analyzing the usage preference of each vehicle-mounted application to generate a usage preference value of each vehicle-mounted application includes: Analyzing the processed multi-dimensional data based on the pre-built target deep learning model to generate a usage preference vector for each vehicle-mounted application; The usage preference vector is used to determine the usage preference value of each vehicle-mounted application.

4. The method according to claim 3, characterized in that The using the usage preference vector to determine the usage preference value of each vehicle-mounted application includes: Determining a recommendation score for each vehicle-mounted application based on a target application feature vector and the usage preference vector in the vehicle-mounted Internet of Vehicles ecosystem; The usage preference value of each vehicle-mounted application is determined according to the recommendation score of each vehicle-mounted application.

5. The method according to claim 1, characterized in that After determining the vehicle application preference analysis result of the vehicle according to the usage preference value, the method further includes: Generating a recommended layout of vehicle-mounted computer applications for the vehicle according to the vehicle-mounted computer application preference analysis result; The recommended layout of the vehicle-mounted computer application is sent to a preset terminal, so that the recommended layout of the vehicle-mounted computer application is displayed on the preset terminal.

6. A device for analyzing vehicle application preference, characterized in that: include: A collection module, used to collect the driving status data of the vehicle and the usage frequency data, usage duration data, switching frequency data and application rating data of each vehicle application in the vehicle by the user; an acquisition module, configured to perform target data processing on the driving state data and at least one of the usage frequency data, the usage duration data, the switching frequency data and the application score data to obtain processed multidimensional data; An analysis module is used to input the processed multidimensional data into a pre-built target deep learning model, analyze the usage preference of each vehicle-mounted application to generate a usage preference value for each vehicle-mounted application, and determine the vehicle-mounted application preference analysis result of the vehicle according to the usage preference value.

7. The device according to claim 6, characterized in that The acquisition module comprises: A first processing unit, configured to perform target denoising processing on at least one of the position information, the driving speed and the road condition information in the driving state data to obtain first processed data; a second processing unit, configured to perform target data cleaning on at least one of the usage frequency data, the usage duration data, the switching frequency data, and the application rating data to obtain second processed data; The first determining unit is used to determine the processed multi-dimensional data using the first processed data and the second processed data.

8. The device according to claim 6, characterized in that The analysis module comprises: an analyzing unit, configured to analyze the processed multi-dimensional data based on the pre-built target deep learning model to generate a usage preference vector for each vehicle-mounted application; The second determining unit is used to determine the usage preference value of each vehicle-mounted application by using the usage preference vector.

9. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for analyzing vehicle-machine application preferences as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for analyzing vehicle computer application preferences as described in any one of claims 1 to 5.

Images