Cross-platform vehicle machine function classification and adaptation optimization method and system

Abstract

The invention relates to the technical field of vehicles, and discloses a cross-platform vehicle machine function classification and adaptation optimization method and system, and the method comprises the steps: obtaining screen hardware parameters of different vehicle types, mapping the screen hardware parameters to the same vector space for clustering, and dividing a vehicle machine into n2 discrete screen hardware levels; aiming at each screen hardware level, designing a corresponding vehicle machine function classification scheme, collecting use behavior data of different users on vehicle machine functions, generating user preference portraits, and generating a personalized function recommendation list in combination with the vehicle machine function classification schemes; for a to-be-adapted target vehicle machine, determining a screen hardware level of the to-be-adapted target vehicle machine, calling a corresponding vehicle machine function classification scheme and a personalized function recommendation list, and adaptively generating a vehicle machine function list conforming to vehicle type characteristics corresponding to the target vehicle machine; according to the invention, the personalized in-vehicle function list can be automatically adaptively generated according to the in-vehicle hardware capability and the user preference, so that the practicability of the in-vehicle function and the user experience are improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and more specifically, to a cross-platform vehicle machine function classification and adaptation optimization method and system. Background Art

[0002] With the rapid development of intelligent connected cars, the functions of in-vehicle infotainment systems are becoming increasingly rich. However, due to the large differences in hardware configurations and screen parameters of different models, the functional adaptation and personalization of the vehicle system are insufficient, making it difficult to meet the diverse needs of users. There have been some related studies on the optimization and adaptation of in-vehicle system functions in the prior art, but there are still some shortcomings.

[0003] The patent application with publication number CN110888578A discloses a vehicle function optimization method, device and vehicle having the same. The method collects the target operation actions of the driver and passengers, identifies the corresponding vehicle functions, and recommends vehicle functions to the driver and passengers based on the operation frequency of the target operation actions. Although this method can recommend functions based on the user's operating habits, it does not take into account the differences in hardware configurations of different models and cannot achieve cross-platform function adaptation. In addition, this method only makes recommendations based on the operation frequency, does not deeply explore the user's preference characteristics, and is not personalized enough.

[0004] The patent application with publication number CN118760376A discloses an adaptive display method, device, vehicle and storage medium for a vehicle-machine interface. This method obtains the usage time and operation mode of the vehicle-mounted software under different driving conditions, builds a list of commonly used software items for different scenarios, and automatically switches the interface according to the driving status of the vehicle. This method can dynamically adjust the vehicle-machine interface according to the driving scenario, thereby improving the user experience. However, this method does not take into account the differences in the vehicle-machine hardware platform. For screens of different sizes and resolutions, the interface layout and interaction methods cannot be uniformly optimized. At the same time, this method mainly focuses on the scene adaptation of information display, and does not involve the classification and personalized recommendation of vehicle-machine functions.

[0005] In summary, existing vehicle machine function optimization and adaptation methods generally have the problems of poor cross-platform uniformity and insufficient personalization. The screen sizes and resolutions of vehicle machines of different models vary greatly, making it difficult to unify the layout and interaction methods of the vehicle machine interface, and difficult to meet the personalized needs of different users. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a cross-platform vehicle-machine function classification and adaptation optimization method and system, which, on the basis of unified vehicle-machine function classification, realizes personalized function recommendation in combination with user portraits, thereby significantly improving the user experience of vehicle-machine human-machine interaction.

[0007] To achieve the above object, the present invention provides the following technical solutions: A cross-platform vehicle machine function classification and adaptation optimization method, comprising: Obtain screen hardware parameters of different vehicle models, map the screen hardware parameters to the same vector space for clustering, and divide the vehicle computer into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Design a corresponding vehicle function classification scheme for each screen hardware level; collect different users' usage behavior data on vehicle functions to generate user preference profiles; combine the user preference profiles and the vehicle function classification scheme to generate a personalized function recommendation list; For the target vehicle computer to be adapted, the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer are obtained, the screen hardware level of the target vehicle computer is determined, and according to the screen hardware level of the target vehicle computer, the corresponding vehicle computer function classification scheme and personalized function recommendation list are retrieved, and a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer is adaptively generated.

[0008] Furthermore, mapping the screen hardware parameters to the same vector space for clustering includes: Preprocess and select features of screen hardware parameters to obtain clustering features; The clustering features are used as coordinate axes of a multidimensional vector space, and the screen hardware parameters of each vehicle computer are represented as an n1-dimensional screen hardware parameter vector in the multidimensional vector space; Taking the screen hardware parameter vector of the vehicle computer as input, the K-means clustering algorithm is used to divide the vehicle computer systems of different models into n2 screen hardware levels.

[0009] Furthermore, the screen hardware parameters are preprocessed and feature selected to obtain cluster features including: Clean the screen hardware parameters to remove outliers and missing values; Normalize the hardware parameters of the screen after cleaning; Eliminate redundant parameters of the normalized screen hardware parameters; construct parameter pairs based on the normalized screen hardware parameters, calculate the Pearson correlation coefficient between the parameter pairs, and for parameter pairs whose absolute value of the correlation coefficient is greater than a preset correlation threshold, eliminate one parameter in the parameter pair and keep the other parameter as the retained parameter; The principal component analysis is performed on the screen hardware parameters after the redundant parameters are eliminated, the variance contribution rate of each principal component is calculated, and the principal component whose variance contribution rate exceeds the preset contribution rate threshold is used as the clustering feature.

[0010] Furthermore, for each screen hardware level, the corresponding vehicle function classification scheme is designed, including: According to the hardware configuration parameters of the vehicle computer, the computing performance index of the vehicle computer is obtained; According to the vehicle computer screen parameters, the display effect index of the vehicle computer is obtained; Calculate the functional complexity index and resource occupancy index of the vehicle computer; According to the computing performance index, display effect index, functional complexity index and resource occupancy index of the vehicle computer, the support vector machine (SVM) algorithm is used to construct a functional classification model, which divides the vehicle computer functions into three levels: basic functions, extended functions and advanced functions.

[0011] Furthermore, the usage behavior data includes the user's demographic characteristics and function usage records; Generating a user preference profile includes: The demographic characteristics and function usage records of each user are constructed into a user behavior feature vector, and the user behavior feature vector is clustered to divide the users into n3 user groups; The average usage frequency and duration of different functions by each user group are counted, a user-function preference matrix is ​​constructed, and a function preference portrait of each user group is generated; the rows of the user-function preference matrix represent different user groups, the columns represent various vehicle computer functions, and the matrix elements represent the degree of preference of a certain user group for a certain vehicle computer function.

[0012] Furthermore, generating a personalized function recommendation list includes: Based on user preference profiles, obtain a list of preferred functions for different user groups; For each screen hardware level, filter the high-frequency preferred functions from the corresponding vehicle machine function classification scheme to obtain a candidate function set; Generate a personalized feature recommendation list from the candidate feature set.

[0013] Furthermore, the obtaining of the preferred function lists of different user groups includes: Sort each row of the user-function preference matrix in descending order to obtain a function ranking list from high to low preference for each user group; Intercept the N preferred functions with the highest preference in each function ranking list to generate a preferred function list for each user group; The occurrence frequency of each preference function in the preference function list is counted, and the preference functions whose occurrence frequency exceeds a preset frequency threshold are marked as high-frequency preference functions to generate a high-frequency preference function set.

[0014] Furthermore, the high-frequency preference functions are screened from the corresponding vehicle machine function classification scheme to obtain a candidate function set including: Include all basic functions and extended functions in the high-frequency preference function set into the candidate function set; Calculate the resource occupation level SE of each high-level function in the high-frequency preference function set; The high-level functions in the high-frequency preference function set whose resource occupancy level SE is lower than a preset resource occupancy level threshold are included in the candidate function set.

[0015] Furthermore, generating a personalized function recommendation list from the candidate function set includes: Obtain demographic characteristics and historical function usage records of the user to be recommended, and construct a user behavior feature vector of the user to be recommended; calculate the similarity between the user behavior feature vector of the user to be recommended and the user behavior feature vectors of each user group, and find the user group with the highest similarity as the user group to which the user to be recommended belongs; Obtain the preferred function list of the user group to which the recommended user belongs, and find the intersection with the candidate function set to obtain the candidate recommended function set; Based on the historical function usage records of the user to be recommended, the historical function preferences of the user to be recommended are obtained, and according to the historical function preferences of the user to be recommended, M functions are selected from the candidate recommendation function set to generate a personalized function recommendation list.

[0016] A cross-platform vehicle machine function classification and adaptation optimization system, which is used to implement the above-mentioned cross-platform vehicle machine function classification and adaptation optimization method, the system includes: Vehicle computer classification module: obtains screen hardware parameters of different vehicle models, maps the screen hardware parameters to the same vector space for clustering, and divides the vehicle computers into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Personalized recommendation module: Design a corresponding vehicle function classification scheme for each screen hardware level; collect different users' usage behavior data on vehicle functions to generate user preference profiles; combine the user preference profiles and the vehicle function classification scheme to generate a personalized function recommendation list; Adaptation module: For the target vehicle computer to be adapted, obtain the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determine the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieve the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generate a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer.

[0017] An electronic device includes a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, the above-mentioned cross-platform vehicle machine function classification and adaptation optimization method is implemented.

[0018] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the above-mentioned cross-platform vehicle machine function classification and adaptation optimization method.

[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention can map different types of vehicle computers to a unified screen hardware hierarchy system by extracting features and performing cluster analysis on vehicle computer screen hardware parameters, providing a basis for subsequent function classification and optimization. This method of hardware parameter mapping can effectively overcome the adaptation difficulties caused by vehicle computer hardware fragmentation.

[0020] The present invention designs a function classification scheme that matches the hardware capabilities for each screen hardware level, and divides the vehicle computer functions into three levels: basic, extended, and advanced. This hierarchical adaptation strategy can maximize the use of vehicle computer hardware performance, reasonably configure functions at each level, and avoid resource waste or insufficient performance.

[0021] The present invention collects user vehicle usage behavior data and generates user function preference portraits using algorithms such as clustering and collaborative filtering. Based on the user portrait, a personalized recommended function list is generated, which can accurately match user needs, provide more intimate vehicle function services, and bring a smoother human-computer interaction experience.

[0022] The cross-platform vehicle machine function classification and adaptation optimization method of the present invention has strong versatility and scalability. For the access of new models, only the screen hardware parameters need to be obtained to automatically match the corresponding screen hardware level, function classification scheme and recommendation list, thus realizing one-stop function adaptation optimization and greatly reducing the manual development cost.

[0023] The cross-platform vehicle function classification and adaptation optimization method of the present invention fully taps the value of user usage data, realizes the trinity of hardware perception, user perception, and function perception, and builds an end-to-end, full-process vehicle personalized function customization optimization platform. This provides a humanized and refined function management and operation means for intelligent networked vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 A method flow chart of a cross-platform vehicle machine function classification and adaptation optimization method in the present invention; Figure 2 A flowchart of a method for mapping screen hardware parameters to the same vector space for clustering in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 3 A flow chart of a method for obtaining clustering features in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 4 A flow chart of a method for designing a vehicle machine function classification scheme in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 5 A flow chart of a method for generating a user preference portrait in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 6 A flowchart of a method for obtaining a list of preferred functions of different user groups in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 7 A flow chart of a method for obtaining a candidate function set in a cross-platform vehicle machine function classification and adaptation optimization method of the present invention; Figure 8 This is a functional module diagram of a cross-platform vehicle machine function classification and adaptation optimization system in the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Example 1 See also Figure 1 As shown, this embodiment provides a cross-platform vehicle machine function classification and adaptation optimization method, including: Step S1000, obtaining screen hardware parameters of different vehicle models, mapping the screen hardware parameters to the same vector space for clustering, and dividing the vehicle computer into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Furthermore, step S1000 includes: Step S1100, obtaining vehicle screen parameters of different vehicle models; the vehicle screen parameters include vehicle screen size parameters, vehicle screen resolution parameters, vehicle screen pixel density parameters and vehicle screen color gamut parameters; Specifically, the car screen size parameters include the screen diagonal length, aspect ratio, etc.; the car screen resolution parameters include the horizontal pixel count and the vertical pixel count; the car screen pixel density parameters are calculated by the screen diagonal length and resolution; the car screen color gamut parameters include the coverage of different color gamut standards such as Adobe RGB (Adobe red, green, blue color space), sRGB (standard red, green, blue color space), and NTSC (color television broadcasting standard established by the National Television System Committee of the United States). Step S1100 comprehensively understands the heterogeneity of different car screens by collecting a large range of car screen parameters of mainstream models on the market.

[0028] The car screen size parameter reflects the physical size of the car screen and is a key factor in determining the UI layout. By measuring the diagonal length of the screen, the approximate size of the screen can be obtained; the screen aspect ratio describes the ratio of the length to the width of the screen, which affects the display of UI elements. Common aspect ratios include 4:3, 16:9, 16:10, etc. The screen resolution indicates the fineness of the screen image. The higher the resolution, the richer the image details that the screen can display. The horizontal and vertical pixel counts of the screen can accurately describe the resolution value. For example, 1280×720 means 1280 pixels in the horizontal direction and 720 pixels in the vertical direction. The pixel density reflects the clarity of the screen, expressed in pixels per inch, recorded as PPI (Pixels Per Inch). PPI can be converted from screen size and resolution. The higher the PPI, the more delicate the display effect. The color gamut parameter characterizes the color expression of the screen. Different color gamut standards have different color coverage. The Adobe RGB color gamut covers a wider color space, especially the green and cyan parts, which are suitable for high-end display needs such as printing; the sRGB color gamut is the most widely used standard, covering most of the colors that the human eye can perceive; the NTSC color gamut is derived from the television broadcast standard and emphasizes the performance of red and yellow. By collecting the coverage data of the car screen under different color gamut standards, the color display capability of the car can be quantitatively evaluated.

[0029] The car screen parameters characterize the characteristics of the car screen from multiple dimensions such as the physical size of the screen, display accuracy, and color performance. There are both quantitative parameter values ​​and qualitative grade descriptions, which provide solid data support for subsequent screen grading. The acquisition of car screen parameters requires a lot of manpower and material resources. It is a long and meticulous process through field visits, field measurements, consulting vehicle manuals, and comparing screen specifications. However, complete and accurate screen parameter data is the basis for cross-platform interface adaptation, so the role of step S1100 is crucial. Collecting comprehensive car screen parameters unifies the reference standards for subsequent adaptation and optimization, avoids analysis blind spots caused by missing data, and is a necessary condition to ensure the adaptation effect.

[0030] Step S1200, obtaining the hardware configuration parameters of the vehicle system of each vehicle model; the hardware configuration parameters include CPU model, memory capacity parameters, GPU model and storage device parameters; Specifically, the CPU model includes parameters such as the number of CPU cores, main frequency, and cache capacity; the memory capacity parameters include the specifications and models of the running memory and storage memory; the GPU model includes parameters such as the number of GPU cores, frequency, and video memory capacity; the storage device parameters include storage capacity, storage medium type (such as eMMC, UFS), read and write speeds, etc.; Step S1200 evaluates the computing performance differences of different vehicle systems by collecting the hardware configurations of the vehicle systems of various models.

[0031] As the central processor of the car computer, the performance parameters of the CPU directly determine the computing speed and concurrent processing capability of the car computer. By comparing the key indicators of different CPU architectures (such as ARM, X86), process technology, number of cores, main frequency, etc., the computing power of the car computer can be quantitatively analyzed. For example, the car computer equipped with Qualcomm Snapdragon 865 has a more advanced 5nm process and more CPU cores than the car computer equipped with Snapdragon 835, and the computing speed can be increased by more than 30%. Memory capacity is another factor that affects the smoothness of the car computer operation. The specifications of the running memory and storage memory can be collected to determine the data processing capability of the car computer. 8GB of LPDDR5 (low-power double data rate fifth-generation memory) running memory can drive multiple large software at the same time, while 256GB of UFS3.1 (universal flash storage 3.1 standard) mobile phones have larger storage space and faster data reading speed. The GPU undertakes the graphics rendering task of the car computer. The model and frequency parameters of the GPU can be collected to judge the graphics processing performance of the car computer. For example, a car computer equipped with Mali-G78 has more GPU cores and higher frequency than Mali-G72, and the graphics rendering speed can be increased by about 50%. The type and capacity of the storage device will also affect the user experience of the car computer. Collecting the storage capacity and media type can analyze the data storage capacity of the car computer. The 256GB UFS storage has more space to store map data and music files, while the reading and writing of eMMC storage is relatively slow, which may cause problems such as jamming.

[0032] The hardware configuration parameters evaluate the performance level of the car computer from multiple dimensions, such as computing, graphics, and storage, and cover the key hardware factors that affect the user experience of the car computer. The collection of hardware parameters requires consulting the product manual of the car computer, disassembling the equipment for testing, and comparing the running data of similar products. The process is relatively complex and professional. However, reliable hardware parameter evaluation data is the basis for ensuring the adaptation effect of the car computer software. Ignoring the differences in hardware performance and carrying out one-size-fits-all software development is bound to cause user experience problems such as freezes and crashes. Collecting comprehensive hardware configuration parameters clarifies the hardware endowments of different car computers and is an important basis for judging the adaptation optimization effect. Only within the scope permitted by hardware conditions can the optimization space of software algorithms be brought into play to adjust the best user experience.

[0033] Step S1300, mapping the screen hardware parameters to the same vector space for clustering, and dividing the vehicle computer into n2 discrete screen hardware levels.

[0034] Furthermore, if Figure 2 As shown, step S1300 includes: Step S1310, preprocessing and feature selection are performed on screen hardware parameters to obtain cluster features; Furthermore, if Figure 3 As shown, step S1310 includes: Step S1311, cleaning the screen hardware parameters to remove obvious abnormal values ​​and missing values; Step S1312, normalizing the hardware parameters of the cleaned screen; Step S1313, removing redundant parameters from the normalized screen hardware parameters; constructing parameter pairs according to the normalized screen hardware parameters, calculating the Pearson correlation coefficient between the parameter pairs, and for parameter pairs whose absolute values ​​of the correlation coefficients are greater than a preset correlation threshold, removing one parameter in the parameter pair and leaving the other parameter as a retained parameter; Step S1314, performing principal component analysis on the screen hardware parameters after removing redundant parameters, calculating the variance contribution rate of each principal component, and taking the principal component whose variance contribution rate exceeds a preset contribution rate threshold as a clustering feature.

[0035] Specifically, outliers are usually caused by measurement errors, data entry errors and other factors. They are far from normal observation values ​​and will interfere with subsequent statistical analysis. For example, the running memory of general car computers is between 1GB and 12GB. If the memory of individual car computers is as high as 128GB, this is likely to be an outlier caused by misrecording and needs to be eliminated. Similarly, the missing values ​​of individual parameters will also affect the overall data quality and need to be estimated or directly eliminated. After completing the processing of outliers and missing values, the validity and cleanliness of the data can be improved. Screen hardware parameters have different dimensions and the values ​​vary greatly. For example, the screen size is measured in inches, generally between 5-15 inches, and the resolution value can reach tens of millions. If the original values ​​are used directly for clustering, the impact of resolution will be much greater than that of screen size. Normalization processing unifies different parameters into similar numerical ranges and gives them similar importance. Common normalization methods include Min-Max normalization and Z-score normalization. The normalized parameters are easier to compare comprehensively.

[0036] Step S1313 removes redundant parameters through correlation analysis. There may be significant linear correlation between some parameters, carrying similar information. For example, resolution is highly correlated with pixel density, and incorporating clustering at the same time will increase the consideration of display accuracy. Redundant parameters will interfere with the clustering effect, prolong the calculation time, and need to be optimized and eliminated. Here, the Pearson correlation coefficient is used to measure the linear correlation between any two parameters. The closer the absolute value of the coefficient is to 1, the stronger the positive or negative correlation is. For parameter pairs that exceed the preset correlation threshold, the two carry similar information, and one of them can be eliminated. Step S1314 further uses principal component analysis to extract clustering features. Principal component analysis is a commonly used dimensionality reduction method that can condense multiple parameters into a few principal components while retaining data features. Each principal component is a linear combination of the original parameters and carries most of the information of the original data. The variance contribution rate of the principal component measures its ability to explain the original data. The higher the contribution rate, the more complete the information is retained. By analyzing the variance contribution rate of each principal component and selecting a few principal components with higher contribution rates as clustering features, the original information can be retained to the maximum extent while reducing the computational complexity.

[0037] For example, assume that five key parameters of 100 car models are collected: screen size, resolution, pixel density, CPU frequency, and memory capacity. Some of the data is shown in Table 1: Table 1 List of some data on the five key parameters of 100 car models

[0038] Execute steps S1311 to S1314 in sequence for this set of data: Step S1311: Clean abnormal values ​​and missing values. The screen size 230 in row 4 is obviously abnormal, so delete this row.

[0039] Step S1312: Min-Max normalization, mapping each column of original data to the interval [0,1].

[0040] Step S1313: Eliminate redundant parameters. Calculate the Pearson correlation coefficients of all parameter pairs, and the results are shown in Table 2: Table 2 Pearson correlation coefficient results data table

[0041] The absolute values ​​of correlation coefficients exceeding 0.9 are: screen size and resolution, screen size and CPU frequency. Resolution and CPU frequency are eliminated, and screen size, pixel density and memory capacity are retained.

[0042] Step S1314: principal component analysis. Perform principal component analysis on the normalized data matrix after removing redundant parameters, and obtain the variance contribution rates of the principal components as 80%, 15%, and 5%, respectively. The cumulative contribution rate of the first two principal components has reached 95%, which is sufficient to represent most of the information of the original data. These two principal components are selected as clustering features.

[0043] Step S1320, using the clustering features obtained in step S1310 as the coordinate axes of the multidimensional vector space, and representing the screen hardware parameters of each vehicle computer as an n1-dimensional screen hardware parameter vector in the multidimensional vector space; For example, assuming that the screen hardware parameters of the vehicle computer still retain 6 key parameters after dimensionality reduction, it can be represented as a 6-dimensional screen hardware parameter vector: ; Among them, pixel_density represents the screen pixel density, color_gamut represents the color gamut coverage, cpu_freq represents the CPU main frequency, gpu_freq represents the GPU frequency, ram_size represents the memory capacity, storage_size represents the storage capacity, and vehicle_vector represents a 6-dimensional screen hardware parameter vector.

[0044] In this multidimensional vector space, the screen hardware parameters of each model correspond to a position coordinate. The specific values ​​of the model in each parameter dimension determine its position in the vector space. By calculating the vector distance between models, the similarity of the car screen hardware parameters can be quantitatively evaluated, providing a mathematical basis for the subsequent hierarchical clustering of car hardware.

[0045] Specifically, by modeling the hardware parameters of the car screen as a vector space, the multi-dimensional heterogeneous parameters of different models can be mapped to a unified feature space for comparison. Each coordinate axis in the feature space represents a key parameter, and the scale of the coordinate axis reflects the numerical value of the parameter. The specific value of the screen hardware parameters of each model determines its position coordinates in the feature space, and the position coordinates reflect the comprehensive performance of the model in each key parameter. In the feature space, the Euclidean distance between models reflects their similarity in hardware performance. The closer the distance, the closer the hardware parameters, and the more similar the adaptation optimization strategies. Therefore, the mapping of the hardware parameters of the car screen to the vector space provides an intuitive geometric representation of the screen hardware differences of different models, and also lays the model foundation for hardware grading using clustering algorithms. Unifying the parameters of multiple dimensions into the same space facilitates the overall evaluation of the hardware performance of the model without the need for independent threshold judgment and weighted summation of each single parameter, avoiding the influence of subjective factors and improving the objectivity and interpretability of hardware grading.

[0046] Step S1330, taking the screen hardware parameter vector of the vehicle computer as input, using the K-means clustering algorithm to divide the vehicle computer systems of different models into n2 screen hardware levels.

[0047] Specifically, K-means clustering divides similar car machines into the same screen hardware level, and there are obvious differences in hardware performance between car machines of different levels. The division of screen hardware levels comprehensively considers key parameters such as screen display, computing performance, and storage capacity, reflecting the overall differences in hardware performance of car models. High-level car machines have better screen display effects, more powerful computing power, and larger storage space, and can adapt to more sophisticated UI interfaces and run more complex functions; while low-level car machines are limited by hardware and need to adapt to simpler UIs and functions to ensure basic experience. K-means clustering makes full use of the distribution characteristics of hardware parameters of multiple models, and through iterative search for the optimal clustering center, the hardware heterogeneity of the model is converted into several discrete levels, which is convenient for subsequent interface and function classification adaptation. Compared with the hardware classification method of single threshold division, K-means clustering can perform adaptive classification according to the distribution of the data itself, which is more objective and flexible. In addition, K-means clustering has the advantages of high computational efficiency and good interpretability. The screen hardware level of each car model is clearly defined, and the hardware differences are intuitively comparable, which is more conducive to the formulation of graded adaptation strategies.

[0048] For example, assume that the car computers are clustered into three screen hardware levels: low-end, medium-end, and high-end. The clustering results may be as follows: Low-end car machine: screen pixel density is less than 200ppi, color gamut coverage is less than 90% sRGB, equipped with low-end CPU / GPU, main frequency is less than 1.5GHz, memory is less than 4GB, storage capacity is less than 32GB. Representative models include some entry-level SUVs, economy cars, etc.

[0049] Mid-range car computers: screen pixel density is between 200-300ppi, color gamut coverage is 90%-95% sRGB, equipped with mid-range CPU / GPU, main frequency is 1.5-2.5GHz, memory is 4-8GB, storage capacity is 32-128GB. Representative models include some mid-range SUVs, compact cars, etc.

[0050] High-end car machine: screen pixel density higher than 300ppi, color gamut coverage higher than 95% sRGB, equipped with flagship CPU / GPU, main frequency higher than 2.5GHz, memory higher than 8GB, storage capacity higher than 128GB. Representative models include some luxury SUVs, flagship sedans, etc.

[0051] For low-end car computers, the UI interface should adopt a simpler graphic design style, reduce visual effects such as gradients, shadows, and transparency, and reduce the complexity of interface drawing; at the same time, simplify the functions of the car computer, retain only core applications such as navigation and music, and do not provide extended functions such as video playback and voice assistants.

[0052] For mid-range car computers, the UI interface can adopt a flat design, with moderate use of visual elements such as gradients and images, but overly complex multi-layer overlays should be avoided; functions should integrate common applications such as navigation, music, and radio, and provide extended functions such as video playback and voice assistant, but the video bit rate and complexity of voice interaction should be limited.

[0053] For high-end car computers, the UI interface can adopt a skeuomorphic design, making extensive use of visual elements such as gradients, shadows, and motion effects to create a refined interface experience; in addition to covering all applications of mid- and low-end car computers, it can also integrate innovative applications exclusive to large-screen car computers, such as KTV, games and other entertainment functions, making full use of high-configuration screens, CPUs and memory to provide the ultimate in-car experience.

[0054] The screen hardware of the car machine is graded through K-means clustering, and targeted UI design and function implementation strategies are formulated in combination with the grade division. While ensuring the user experience, excessive requirements for low-end car machine hardware are avoided, thereby reducing the development cost of adaptation and optimization and improving the delivery efficiency of the car machine system. In addition, the hardware grade division is conducive to car companies to divide the configuration requirements of different models according to positioning and cost, and hardware procurement can also adapt to business strategies.

[0055] Step S2000, designing a corresponding vehicle machine function classification scheme for each screen hardware level; collecting usage behavior data of different users on vehicle machine functions to generate user preference portraits; combining the user preference portraits and the vehicle machine function classification scheme to generate a personalized function recommendation list; Furthermore, step S2000 includes: Step S2100, designing a corresponding vehicle machine function classification scheme for each screen hardware level; the vehicle machine function classification scheme includes three levels: basic functions, extended functions, and advanced functions; Furthermore, if Figure 4 As shown, step S2100 includes: Step S2110, obtaining computing performance indexes of vehicle computers with different screen hardware levels according to hardware configuration parameters of vehicle computers with different screen hardware levels; ; in: , the computing performance index of the car computer under the screen hardware level L.

[0056] , the baseline adjustment coefficient of screen hardware level L; Function: Used to balance the hardware configuration differences of different screen hardware levels. It takes different values ​​for different screen hardware levels, and the value range is adjusted according to the overall performance of the vehicle screen hardware level. The preferred value range is generally between 1.0 and 2.5. The higher the level, the better. The bigger. Ensure that the computing performance index of different screen hardware levels will not fluctuate too much, and appropriately amplify the performance of high-level car computers to highlight the level differences.

[0057] How to obtain: The optimal value can be obtained through regression analysis of historical performance data or machine learning model optimization.

[0058] , the average CPU performance of screen hardware level L; effect: Indicates the average CPU performance of the vehicle computer at screen hardware level L, which is used to evaluate the computing and processing capabilities of the vehicle computer at this level. CPU performance is the core of vehicle computer operations and affects the computing speed and response time. The level directly determines The computing core capabilities in .

[0059] Acquisition method: Get the CPU performance data of all car computers within the screen hardware level L and take the average.

[0060] , CPU base coefficient; effect: Controlling CPU performance average The logarithmic amplification effect, balance Value pair Balance the performance of different CPU levels to prevent low-performance CPUs from causing The value is too low. The preferred value range is generally between 0.1 and 0.8, and the specific value is adjusted according to the CPU performance range of the screen hardware level L; Acquisition method: Analyze the CPU performance data distribution of each hardware level to ensure Time Calculations maintain moderate impact.

[0061] , CPU index adjustment factor; effect: It is an exponential adjustment factor that affects CPU performance. Amplifies as the CPU performance index changes. Controls CPU performance The change rate ensures that the CPU is high and low The impact of CPU on performance index is more reasonable and can be distinguished into different levels. The preferred value range is generally between 0.8 and 1.5, and the value depends on the impact of CPU on performance index; Acquisition method: It can be obtained by regression fitting of CPU performance data of different levels. More in line with the actual performance of the car.

[0062] , memory capacity influence coefficient; effect: Is the memory capacity The influence coefficient of The changes can be more accurately reflected in Adjust the memory capacity to The contribution of The optimal value range is recommended to be between 0.2 and 0.6, and the higher the level, the larger the value; Acquisition method: Determined by regression fitting of historical data based on the impact of memory on multi-task performance.

[0063] , the mean memory capacity of screen hardware level L; effect: Reflects the multi-tasking capability of the car computer. The larger the memory, the more applications and tasks the car computer can support, and the greater the performance improvement effect. Acquisition method: average the memory data of all vehicle computers in screen hardware level L.

[0064] , the benchmark denominator coefficient; effect: Adjust GPU performance and storage speed The impact of GPU and storage speed is balanced to ensure that both The preferred value range is generally between 0.5 and 1.2; Acquisition method: Observe the distribution of GPU and storage performance data of different levels and fit the optimal value.

[0065] , storage adjustment coefficient; effect: Used to adjust the storage speed to ensure that slow storage devices do not significantly reduce , so that The calculation remains stable in low-level devices. The preferred value range is between 0.1 and 1.0, and higher values ​​can be taken at higher levels; Acquisition method: through historical storage data fitting or empirical judgment.

[0066] , the average GPU performance of the car machine under the screen hardware level L; effect: It is the average GPU performance of the vehicle at screen hardware level L, and is used to evaluate image processing and rendering capabilities. Higher GPU performance reflects better graphics and video processing capabilities, improving the overall performance index; Acquisition method: Calculate the average of the GPU performance data of all car computers in screen hardware level L.

[0067] , the average storage speed of the car computer under the screen hardware level L; Function: Storage speed affects application startup time and data processing speed. The higher the storage performance, the faster the application startup time and data processing speed. The more significant the improvement; Acquisition method: Average the speed data stored in the vehicle computer in the screen hardware level L.

[0068] , resolution influence coefficient; effect: Used to adjust the average screen resolution right The preferred value range is generally between 0.1 and 0.6, and the specific value is related to the screen level and resolution requirements; Acquisition method: Obtained through data analysis of the impact of resolution on performance.

[0069] , mean screen resolution; effect: Reflects the visual display quality of the vehicle computer. The higher the resolution, the better the visual experience, but also the higher the performance requirements; Acquisition method: calculate the average of the vehicle computer resolution data in the screen hardware level L.

[0070] This formula maps the hardware configuration parameters of different screen hardware levels L to In this way, the computing performance index of each level can reasonably reflect the comprehensive performance of its hardware. , By adjusting the parameters such as The increase was obvious, while the low-level equipment remained relatively stable. .

[0071] Step S2120, obtaining the display effect index of the vehicle computers of different screen hardware levels according to the vehicle computer screen parameters of the vehicle computers of different screen hardware levels ; ; in: Indicates the number of samples at the current screen hardware level; Indicates Car machine samples; Indicates The weighting factor of the car machine samples is used to control the impact of different screen hardware levels; Indicates The pixel density parameter of the vehicle machine samples; Indicates Screen resolution parameters of vehicle machine samples; Indicates Screen size parameters of car machine samples; Indicates Screen color gamut coverage parameters of vehicle machine samples; It is the adjustment parameter of pixel density, reflecting the weight of pixel density in the display effect index; It is the adjustment parameter of resolution, reflecting the weight of resolution in the display effect index; It is the adjustment parameter of the screen color gamut coverage; it reflects the weight of the color gamut in the display effect index; It is the adjustment angle of the color gamut weight, which is used to control the relative contribution of display color quality under different screen hardware levels; is a very small positive number (such as ), used to avoid the case where the denominator is zero.

[0072] Weighting Factor It can be set according to the representativeness of each car computer sample at the current level, mainly defined by the sales share or extensive application of the current car computer in a certain level, to ensure that the representative data in the sample has a greater impact on the display effect index.

[0073] Pixel density Reflects the clarity of the screen display, which can be calculated through the resolution and size of the screen. This parameter affects the visual clarity part of the display effect index.

[0074] Screen resolution , used to evaluate the maximum resolution that the screen can display, usually represented by the physical resolution of the screen (such as 1920×1080). This parameter determines the level of detail of the displayed image and affects the clarity of the display effect.

[0075] Screen size Indicates the physical size of the screen, usually in inches. Screen size affects the user's visual experience. A larger screen will reduce the display density at the same resolution, so the denominator is used in the formula to balance the display effect.

[0076] Color gamut coverage It is used to evaluate the color performance of the screen. The color gamut range covered by the screen is usually expressed as a percentage (such as the percentage coverage of sRGB and Adobe RGB). The color gamut coverage determines the richness and authenticity of the screen colors, so the color gamut is introduced into the formula. Function to simulate the sensitivity of the human eye to different levels of color display quality.

[0077] Adjustment parameters These parameters are used to adjust the weight of different display factors on the display effect index to adapt to the differences in screens of different hardware levels. By adjusting the values ​​of these parameters, you can provide personalized weight configurations for different levels of screen hardware.

[0078] The pixel density in this formula When increasing, It will rise accordingly, but the effect will gradually become flat, which is in line with the visual saturation effect of the human eye on high pixel density screens.

[0079] Resolution As it increases, the display effect index will increase linearly, indicating that at the same screen size, the higher the resolution, the better the display effect.

[0080] Screen size When it increases, the display effect index will decrease slightly, because a larger screen will dilute the display density, thus affecting display clarity.

[0081] Color gamut coverage When it increases, the increase of the display effect index has a cyclical trend, simulating the advantage of color quality in high color gamut coverage.

[0082] Through the nonlinear adjustment of this formula, the display effect index can comprehensively reflect the differences between different screen levels and the visual performance of specific screen parameters, so as to more accurately evaluate the display effect of the screen hardware.

[0083] The formula combines multiple parameters such as resolution, pixel density, screen size and color gamut coverage, and achieves clustering in the same vector space in a weighted and mapped manner, thereby calculating the corresponding display effect index for different levels of screen hardware. By setting different weighting factors for each level And adjust the parameters The value of can make the formula flexibly applied in different levels, reflecting the differences of different levels. At the same time, through the nonlinear processing of dependent variables (such as resolution, size, color gamut), the expressiveness of the display effect index is further enhanced, so that it can be sampled and averaged in the same level, while maintaining sufficient differentiation in different levels.

[0084] Step S2130, for each screen hardware level, the vehicle computer functions are divided into three levels: basic functions, extended functions, and advanced functions according to the computing performance index and display effect index of the vehicle computer, and combined with the functional complexity and resource occupancy level.

[0085] Further, step S2130 includes: Step S2131, for each screen hardware level, calculating the function complexity index of the vehicle computer; Step S2132, for each screen hardware level, calculating the resource occupation index of the vehicle computer; Step S2133, for each screen hardware level, according to the computing performance index, display effect index, functional complexity index and resource occupancy index of the vehicle computer, the support vector machine (SVM) algorithm is used to construct a function classification model, and the vehicle computer functions are divided into three levels: basic functions, extended functions and advanced functions.

[0086] Specifically, in step S2131, the complexity index characterizes the difficulty of implementing the vehicle computer function and the logic complexity; the complexity index is obtained by analyzing the code volume of the function, the number of APIs called, the types of third-party libraries involved and other indicators. The higher the index, the more complex the function implementation. The amount of code contained in the function is an intuitive basis for complexity evaluation. The larger the amount of code, the more business logic the function involves and the more complex the interaction between modules. In addition, the number of system APIs and third-party libraries called by the function also reflects its complexity from the side. The more APIs and libraries called, the more system resources may need to be coordinated for the function, and the more complicated the logic flow is. In the actual evaluation process, code static analysis tools can be used to count the number of code lines, API calls, number of referenced libraries, etc. of each functional module, and set empirical thresholds to quantify the complexity index.

[0087] In step S2132, the resource occupancy index characterizes the level of consumption of hardware resources when the vehicle machine function is running; the resource occupancy index is obtained by weighted average of indicators such as CPU occupancy, memory occupancy, storage space occupancy, and network traffic consumption when the function is running. The higher the index, the more resource consumption. The CPU occupancy rate reflects the level of occupancy of processor resources when the function is running. The higher the occupancy rate, the more computing-intensive tasks the function contains. The memory occupancy rate characterizes the consumption of memory resources when the function is running. The larger the occupancy rate, the more data may need to be processed. The storage space occupancy rate measures the degree of dependence of the function on local storage resources. The larger the occupancy rate, the more user data or cached data may need to be saved locally. The network traffic consumption reflects the occupancy intensity of the function on data transmission resources. The higher the traffic rate, the more frequent network data interaction may be required for the function. In actual evaluation, various resource consumption indicators can be collected in real time when the function is running through performance analysis tools, and the weighted average is calculated based on the empirical weights to finally obtain the resource occupancy index.

[0088] Step S2133 uses the support vector machine (SVM) algorithm to train a functional grading model, which realizes the automatic stratification of vehicle functions. The computing performance index, display effect index, complexity index, and resource occupancy index are used as the input features of the model, and the hardware adaptation difficulty of different functions can be comprehensively evaluated from the two perspectives of hardware adaptation and software implementation. Using the functional level as the output label of the model, the target adaptation level of different functions can be clearly defined, and the corresponding relationship between different indexes and functional levels can be learned through the training sample set. Basic functions usually include basic operations of vehicle computers, such as power on and off, brightness adjustment, time display, etc., with low requirements for hardware adaptation; extended functions usually include in-vehicle navigation, multimedia playback, Bluetooth phone, etc., with medium requirements for hardware adaptation; advanced functions may include voice assistants, mobile phone projection, driving assistance, etc., with high requirements for hardware adaptation. Constructing a suitable training sample set is crucial for training high-quality grading models. The training samples need to cover different levels of vehicle hardware and various types of functions, and professionals should annotate the functional level of each sample. The SVM model builds an optimal classification plane by learning the mapping relationship from features to labels in the sample, which can correctly divide different functions into target levels.

[0089] A major advantage of the function grading model is its objectivity and efficiency. Traditional function grading mainly relies on expert experience, which is subjective, time-consuming and inefficient. The trained grading model can automatically infer the target level of adaptation based on the hardware resource consumption of the function, greatly improving the grading efficiency. The grading results can directly guide the subsequent interface and interaction optimization. Advanced functions can adopt more sophisticated UI design and interaction methods, while low-level functions should try to simplify UI elements and interaction steps.

[0090] For example, assuming that a certain car machine belongs to a high-end model, the model input features of its multimedia playback function are as follows: Computing performance index: 0.8 (can run HD video decoding smoothly); Display effect index: 0.9 (display with high contrast and wide color gamut); Complexity index: 0.6 (logics such as audio and video decoding and subtitle rendering are relatively complex); Resource usage index: 0.7 (video decoding and audio playback occupy more CPU and memory resources); The above index is input into the trained function classification model, and the function level output by the model is extended function. This shows that multimedia playback is a medium-difficulty function for high-end car computers. In terms of interface design, UI elements can be appropriately enriched to provide extended functions such as cover browsing and lyrics synchronization. In terms of interaction design, rich interaction methods such as voice control and gesture operation can also be introduced to improve user experience.

[0091] For example, assuming that a certain car machine belongs to a low-end model, the model input features of its navigation function are as follows: Computing performance index: 0.4 (limited processor performance, difficult to run complex navigation algorithms smoothly); Display effect index: 0.5 (display size and resolution are low); Complexity index: 0.8 (path planning, map rendering and other algorithm logic are very complex); Resource usage index: 0.9 (real-time navigation will occupy a lot of CPU, memory and storage resources for a long time); The output level of the navigation function for low-end car computers is an advanced function, which means that the navigation function may exceed the hardware capabilities of the car computer and needs to be greatly simplified. In UI design, the vector elements of the map should be removed as much as possible, only the necessary road lines should be retained, and a large number of 3D building models should be avoided. The operation process should also be simplified in terms of interaction, and quick functions such as one-key navigation should be provided to reduce the user's operating burden. At the same time, the navigation algorithm should be optimized in a targeted manner to provide basic path guidance functions while minimizing resource consumption.

[0092] Another major advantage of the functional grading model is its interpretability. Each grading decision can be mapped to a specific hardware adaptation index. Engineers can analyze the index composition, quickly locate possible hardware bottlenecks, and carry out targeted optimization. During the model training process, the importance of different hardware indicators to grading decisions can also be evaluated to further guide hardware selection and algorithm optimization. In the long run, the grading model can also be continuously optimized and updated, and continuous learning can be achieved by introducing more actual functional adaptation cases, and the grading accuracy can be improved through continuous iteration.

[0093] In summary, steps S2131 to S2133 quantitatively evaluate the difficulty of adapting different car machine functions through software and hardware indicators such as complexity index and resource occupancy index, and use the SVM algorithm to build a grading model, automatically dividing the functions into three levels: basic, extended, and advanced, greatly improving the pertinence and efficiency of cross-platform adaptation. This quantitative grading method overcomes the subjectivity of traditional empirical grading, and the decision-making basis is clear and transparent, and the optimization guidelines are clear and traceable. The grading results can directly guide the hierarchical optimization of interfaces, interactions, and algorithms. Through the differentiation of adaptation strategies, the hardware potential of different levels of car machines can be fully utilized to avoid one-size-fits-all over-design or simplistic design. It can be foreseen that this method will significantly improve the adaptation efficiency and quality of car machine software, and provide solid technical support for cross-platform development of car machines.

[0094] Step S2100 designs a differentiated function classification scheme based on the screen hardware level of the vehicle computer, divides the vehicle computer functions into three levels: basic functions, extended functions, and advanced functions, to achieve refined management of vehicle computer functions and resource allocation optimization. The computing performance index comprehensively considers hardware configuration parameters such as CPU frequency, number of cores, GPU frequency, memory capacity, storage capacity, etc., quantitatively evaluates the computing power of different levels of vehicle computers, and is an important decision-making basis for function classification. The display effect index comprehensively measures display-related parameters such as screen size, resolution, pixel density, color gamut, etc., intuitively reflects the differences in visual experience of different levels of vehicle computers, and affects design decisions in many aspects such as UI layout, motion effects, and color matching.

[0095] For car computers of different hardware levels, according to their computing performance index and display effect index, and taking into account the complexity of the function itself and the resource usage, the car computer functions are divided into three levels: basic functions, extended functions and advanced functions. Specifically, they include: Basic functions: These are the core functions that ensure the basic use of the car computer. They have low resource usage, low computational complexity, and a simple interface, such as radios and driving recorders. All levels of car computers must have basic functions.

[0096] Extended functions: These are extended functions that enhance the user experience of the vehicle computer. They have moderate resource usage and computational complexity and a beautiful interface, such as voice assistants, online music, etc. Mid- to high-end vehicle computers need to have extended functions.

[0097] Advanced functions: These are cutting-edge functions that demonstrate the top performance of the vehicle computer. They occupy large resources, have high computational complexity, and have a cool interface, such as AR real-scene navigation and gesture control. Only high-level vehicle computers need to adapt to advanced functions.

[0098] According to the hardware level and resource endowment of the car machine, functions of different levels are reasonably arranged to avoid resource waste and fluency problems while meeting the basic needs of users. Through differentiated function management mechanisms, high-end car machines and low-end car machines each play their hardware advantages and provide function options that meet their positioning, so that users can get a user experience that matches the hardware strength. In addition, function classification facilitates developers and designers to formulate targeted adaptation strategies. High-end car machines focus on experience optimization, and low-end car machines focus on performance optimization, which greatly improves the efficiency of cross-platform adaptation. Function classification uses a combination of qualitative and quantitative methods to reasonably divide the three levels of basic, extended, and advanced. It is not simply piling up the number of functions, but is based on hardware performance and user experience. It takes into account both the complexity of the function itself and the carrying capacity of the hardware, and achieves a balance between adaptation effect, R&D efficiency, and cost investment. It is a practical and feasible hierarchical adaptation solution.

[0099] Step S2200, collecting usage behavior data of different users on vehicle computer functions to generate user preference profiles; Furthermore, if Figure 5 As shown, step S2200 includes: Step S2210, collecting usage behavior data of different users on vehicle computer functions; the usage behavior data includes user demographic characteristics and function usage records; Specifically, the demographic characteristics of users include gender, age, occupation, income level, driving experience, etc.; the function usage record includes the frequency and duration of the user's use of various functions on the vehicle computer. Step S2210 comprehensively records the user's interaction with the vehicle computer functions through multiple channels such as the user behavior collection software toolkit (SDK), questionnaire survey, and vehicle computer log. Demographic information helps to divide users into different groups, and usage records reflect the differences in function preferences among different groups of people.

[0100] For example, some user behavior data collected by a certain car machine is shown in Table 3: Table 3 User behavior data table User ID gender Age Group Navigation usage times Music playback time Bluetooth call duration U1001 male 35-44 years old 120 times / month 10 hours / month 6 hours / month U1002 female 25-34 years old 90 times / month 30 hours / month 1 hour / month U1003 male 45-54 years old 60 times / month 5 hours / month 18 hours / month Based on demographic characteristics, it can be preliminarily judged that U1001 and U1003 may be middle-aged male business people, for whom navigation and Bluetooth phones are just needs; U1002 may be a young woman who is more concerned about entertainment functions such as music playback.

[0101] Step S2220, constructing the demographic characteristics and function usage records of each user into a user behavior feature vector, clustering the user behavior feature vector, and dividing the users into n3 user groups; Specifically, the clustering takes into account demographic similarity and usage behavior similarity, and groups similar users into the same group. The final n3 groups have significant differences in demographic characteristics and usage preferences.

[0102] For example, K-means clustering is performed on the user behavior feature vector. When the number of clusters is 3, the clustering effect is better, and the users are divided into three user groups: User group 1: mainly male users, mostly over 40 years old, mostly private owners, corporate managers, etc., who prefer navigation and Bluetooth phone functions. It is speculated that this type of users have frequent travel and call needs.

[0103] User group 2: mainly female users, mostly under 40 years old, mostly white-collar workers and students, preferring entertainment functions such as music and video. It is speculated that this type of users has stronger entertainment needs.

[0104] User group 3: mainly taxi drivers, who use cars for work and are mostly male. They use the car computer functions for a long time, and their usage scenarios are mainly productivity functions such as window messages and order management.

[0105] Step S2230, statistics the average usage frequency and duration of different functions by each user group, construct a user-function preference matrix, and generate a function preference portrait of each user group; the rows of the user-function preference matrix represent different user groups, the columns represent various vehicle functions, and the matrix elements represent the degree of preference of a certain user group for a certain vehicle function.

[0106] Specifically, the degree of preference is measured based on indicators such as the average frequency and duration of use of a function by a certain group. The row vector represents the overall preference of a user group for each function, which is called the function preference profile of the group.

[0107] Exemplarily, according to the clustering result of step S2220, a user-function preference matrix is ​​constructed as shown in Table 4: Table 4 User-function preference matrix navigation music video Bluetooth Phone Voice Assistant User Group 1 0.8 0.3 0.2 0.6 0.5 User Group 2 0.4 0.9 0.8 0.2 0.7 User Group 3 0.7 0.2 0.1 0.8 0.6 The matrix elements take values ​​between 0 and 1, with larger values ​​representing stronger preferences.

[0108] From the matrix, we can see that user group 1 prefers navigation and Bluetooth phones, and is a business-oriented user; user group 2 prefers music and video, and is an entertainment-oriented user; user group 3 prefers navigation, Bluetooth phones, and voice assistants, and is a productivity-oriented user. This is consistent with the previous speculation on group characteristics. The row vector of each group is its functional preference portrait, which directly reflects the differences in the functional needs of different groups of people in the car machine.

[0109] The data collection of user behavior on vehicle-mounted computer functions is the data basis for achieving hierarchical adaptation of functions. Only by deeply understanding the behavioral characteristics and usage demands of different user groups can we determine which functions are more important to specific groups of people and should be adapted and optimized first. User behavior collection requires comprehensive and detailed recording of the real interaction between users and vehicle-mounted computers, involving multiple links such as data embedding, data cleaning, and privacy protection, which requires high data engineering capabilities. Once user behavior big data accumulates to a certain level, it will form a key decision-making support for function optimization and adaptation.

[0110] Clustering is used to automatically divide users into multiple groups, which makes full use of the distribution characteristics of the data itself, can reveal the inherent laws of population differences, and is more objective and reliable than subjective grouping. In addition, compared with simple rule division, clustering can better balance the importance of each feature, integrate population attributes and behavioral preferences, characterize the multi-dimensional characteristics of the group, and make the group division more comprehensive and three-dimensional. The clustering method uses the classic K-means algorithm, which has fast computing speed and good interpretability, and can quickly converge to the optimal cluster among millions of users.

[0111] On the basis of user grouping, the user-function preference matrix is ​​further introduced to quantify the function preferences of the groups. The preference matrix intuitively presents the preference differences of different groups on different functions, and vividly depicts the overall preference portrait of the group. For example, it can be clearly seen from the matrix that user groups 1 and 3 have strong preferences for business functions such as navigation and Bluetooth phone, while user group 2 prefers entertainment functions such as music and video. These preference portraits will provide direct optimization ideas for subsequent function classification adaptation and guide personalized function recommendations for different groups.

[0112] Step S2200 proposes an automated, quantitative user function preference analysis method. This method uses demographic information and usage records as data basis, unsupervised clustering methods and matrix analysis as main tools, and can automatically extract user groups and function preference portraits from large-scale, small-grained user behavior data to infer the functional demand trends of the population. Compared with traditional subjective grouping and qualitative description, this method has higher intelligence and computational efficiency, and the resulting group portrait is more accurate and objective, which can greatly improve the pertinence and effectiveness of vehicle-machine function adaptation.

[0113] In summary, step S2200 collects comprehensive user behavior data, and automatically generates user group division and function preference portraits through clustering and matrix analysis, which are important data and model support for hierarchical adaptation of vehicle functions. In the subsequent function weight calculation and hierarchical adaptation, the user preference information obtained here will be fully utilized to judge the importance of different functions from a human perspective, formulate personalized adaptation strategies for different groups of people, and maximize the human-computer interaction experience of the vehicle.

[0114] Step S2300, for each screen hardware level, combining the user preference portrait and the vehicle machine function classification scheme, generates a personalized function recommendation list.

[0115] Furthermore, step S2300 includes: Step S2310, based on the user preference portrait, obtaining a list of preferred functions of different user groups; Furthermore, if Figure 6 As shown, step S2310 includes: Step S2311, sorting each row of the user-function preference matrix in descending order to obtain a function ranking list of each user group's preference from high to low; Step S2312, intercepting the N preferred functions with the highest preference degree in each function ranking list, and generating a preferred function list for each user group; Step S2313 , counting the occurrence frequency of each preference function in the preference function list, marking the preference functions whose occurrence frequency exceeds a preset frequency threshold as high-frequency preference functions, and generating a high-frequency preference function set.

[0116] Specifically, step S2311 can obtain a function ranking list of each user group from high to low preference by sorting the row vectors of the user-function preference matrix in descending order. For example, for the jth user group, if the preference for car navigation is 0.8, the preference for car music is 0.6, and the preference for car phone is 0.4, then the preferred function ranking is: navigation> music> phone. Step S2312 further intercepts the first N bits of the sorting list to obtain the N functions with the highest preference for each group. N can be flexibly set according to actual needs, usually 5-10. The intercepted result is called a preference function list, which reflects the user's core function preference. On this basis, step S2313 counts the frequency of occurrence of each function in the preference function lists of different groups. Preference functions whose frequency of occurrence exceeds the preset frequency threshold can be regarded as high-frequency preference functions under the hardware level of the screen. These functions are highly consistent with the actual needs of most users and deserve key adaptation and optimization. Through the analysis of preferred functions, we have clarified the actual preferences of different user groups for car machine functions, which is conducive to matching user demands in personalized recommendations and improving the accuracy of recommendations. The frequency statistics also reveal the common preferences of mainstream user groups. These preferred functions should be given priority in interface layout and resource allocation to meet user needs to the greatest extent and optimize user experience.

[0117] Step S2320, for each screen hardware level, filter the high-frequency preference functions from the corresponding vehicle machine function classification scheme to obtain a candidate function set; Furthermore, if Figure 7 As shown, step S2320 includes: Step S2321, including all basic functions and extended functions in the high-frequency preference function set into the candidate function set; Step S2322, calculating the resource occupation level SE of each high-level function in the high-frequency preference function set; Step S2323: The advanced functions in the high-frequency preference function set whose resource occupation level SE is lower than a preset resource occupation level threshold are included in the candidate function set.

[0118] Specifically, step S2320 selects high-frequency preference functions from the corresponding vehicle function classification scheme for different screen hardware levels to form a candidate function set. The so-called high-frequency preference functions refer to vehicle functions that are frequently used and popular among specific user groups. Different hardware levels are adapted to different high-frequency preference functions to achieve differentiated function recommendations.

[0119] The generation of candidate function sets follows two principles: First, the basic functions and extended functions of the car computer must be included in the candidate set unconditionally. Basic functions are core functions that ensure the normal use of the car computer, such as navigation, Bluetooth phone, etc., which are rigid needs and the adaptation surface must be wide enough. Extended functions are the characteristic functions of car computer manufacturers, such as voice assistants, mobile phone interconnection, etc., which are differentiated selling points and must also be included in the candidate set. Secondly, the inclusion of advanced functions in the candidate set needs to consider the constraints of hardware resources. Advanced functions usually have high requirements for hardware resources such as computing, storage, and display. Only when the resource occupancy level is lower than the hardware carrying capacity can they run smoothly.

[0120] The screening process of high-frequency preference functions to candidate function sets is as follows: Step S2321, directly include basic functions and extended functions into the candidate set; Step S2322, calculate the resource occupancy level SE of each advanced function. SE can be comprehensively evaluated by indicators such as the number of function calls, memory usage, and frame rate; Step S2323, include advanced functions whose SE is lower than the resource occupancy level threshold into the candidate set. The setting of the threshold needs to take into account both the function quality and the hardware level, to ensure the experience of the recommended function and to prevent hardware overload.

[0121] For example, assume that the high-frequency preference functions of a mid-range car machine include: navigation (basic), Bluetooth phone (basic), smart voice (extended), video player (advanced), 3D game (advanced). Navigation, Bluetooth phone, and smart voice enter the candidate set directly. The SE evaluation of the video player is medium to low, and the SE evaluation of the 3D game is medium to high. If the resource occupancy threshold of the mid-range car machine is set to medium, the video player can be included in the candidate set, and the 3D game is screened out because the SE exceeds the standard.

[0122] The generation of candidate function sets integrates the usage preferences and hardware adaptation capabilities of different user groups. It includes both general functions for all users and exclusive functions for segmented users. It also eliminates necessary hardware bottlenecks and initially realizes hierarchical adaptation of functions. Compared with treating all high-frequency functions equally, car computers of different hardware levels can obtain more targeted function combinations under the candidate set method, which improves the personalization of car computer functions.

[0123] Step S2330: Generate a personalized function recommendation list from the candidate function set.

[0124] Further, step S2330 includes: Step S2331, obtaining demographic characteristics and historical function usage records of the user to be recommended, constructing a user behavior feature vector of the user to be recommended; calculating the similarity between the user behavior feature vector of the user to be recommended and the user behavior feature vectors of each user group, and finding the user group with the highest similarity as the user group to which the user to be recommended belongs; Step S2332, obtaining a list of preferred functions of the user group to which the user to be recommended belongs, and finding the intersection with the candidate function set to obtain a candidate recommended function set; Step S2333: Based on the historical function usage record of the user to be recommended, the historical function preferences of the user to be recommended are obtained, and according to the historical function preferences of the user to be recommended, M functions are selected from the candidate recommended function set to generate a personalized function recommendation list.

[0125] Specifically, step S2330 generates a personalized function recommendation list from the candidate function set, achieving the effect of further matching the hierarchically adapted candidate functions to individual users. Personalized recommendation is based on user portraits and user historical behaviors, mining user group characteristics and individual preferences, and finding the top functions in the candidate function set that best fit the current user.

[0126] The process of generating a personalized function recommendation list is as follows: Step S2331, obtain the demographic characteristics (such as gender, age, occupation) and historical function usage records (such as usage duration and frequency) of the user to be recommended, and construct a feature vector that reflects the user's behavior characteristics. Based on the similarity matching between the feature vector and each user group, find the group to which the current user is most likely to belong; Step S2332, extract the preferred function list of the group to which the current user belongs, and intersect it with the candidate function set to obtain a candidate recommendation set that takes into account group preferences and hardware adaptation; Step S2333, based on the candidate recommendation set, combined with the historical preferences of the current user, select the M head functions with the highest degree of fit to generate a personalized recommendation list.

[0127] The user behavior feature vector encodes the user's static attributes and dynamic behaviors into a unified mathematical representation, which facilitates the calculation of the similarity between users. Commonly used algorithms for matching the similarity between users and groups include cosine similarity, Euclidean distance, Pearson correlation coefficient, etc. The intersection operation further highlights the hardware adaptability of candidate functions on the basis of respecting group preferences. On the one hand, the fusion of historical preferences draws on the user's previous choices, and on the other hand, it can also mine the implicit preferences of similar users through algorithms such as collaborative filtering. The length M of the final recommendation list needs to balance the accuracy and diversity of the recommendations, and generally considers dynamic adjustments to business scenarios.

[0128] For example, a certain SUV model is equipped with the aforementioned mid-range car machine, and its high-frequency preferred functions are screened in step S2320 to obtain a candidate function set {navigation, Bluetooth phone, smart voice, video player}. A young male user uses the car machine for the first time, and the system obtains his characteristics: male, 26 years old, programmer, no historical usage record, and matches him to the "young male technical crowd" based on this. The preferred functions of this crowd are {navigation, smart voice, video player, 3D game}, and the candidate set is intersected to obtain the candidate recommendation set {navigation, smart voice, video player}. Finally, based on the choices of similar people, a recommendation list of 2 functions is generated for him {navigation, smart voice}.

[0129] On the basis of hierarchical adaptation, personalized function recommendation further connects group preferences with personal characteristics and provides a tailor-made function combination. Compared with mass recommendations, personalized recommendations allow each user to obtain exclusive functions that better meet their needs, without having to passively accept redundant functions. The utilization efficiency of vehicle computer resources and user experience have been improved, and the matching degree and stickiness of functions have been further strengthened.

[0130] In summary, filtering high-frequency preference functions as candidate function sets and generating a personalized function recommendation list based on user characteristics is an effective way to achieve hierarchical adaptation of vehicle computer functions and enhance the personalized experience of vehicle computers. This method fully explores the differences between different user groups and different vehicle computer hardware, and realizes the leap from unified adaptation to hierarchical adaptation to personalized adaptation of vehicle computer functions, allowing vehicle computers to match user needs more intelligently. Of course, this method is not achieved overnight. It is necessary to continuously accumulate user feedback in actual applications, dynamically optimize adaptation strategies and recommendation algorithms, in order to find the best balance between hardware utilization and user experience.

[0131] Step S3000, for the target vehicle computer to be adapted, obtain the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determine the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieve the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generate a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer.

[0132] Specifically, the vehicle screen parameters of the target vehicle to be adapted are obtained, including screen size, resolution, pixel density, color gamut, etc. These parameters can be obtained by consulting the vehicle hardware manual, disassembling equipment for measurement, or cooperating with car companies. The hardware configuration parameters of the target vehicle are obtained, including CPU model, main frequency, memory capacity, GPU model, storage capacity, etc. These parameters reflect the computing and storage performance of the vehicle. They can be collected through vehicle system information, running software, etc. The screen and hardware parameters of the target vehicle are preprocessed by feature extraction, normalization, etc. according to the method of step S1300 to obtain a multi-dimensional screen hardware parameter vector of the target vehicle. The preprocessing process here is completely consistent with step S1310-step S1320, in order to ensure that the target vehicle parameters and the vehicle model clustering samples are in the same vector space. The Euclidean distance between the screen hardware parameter vector of the target vehicle and each cluster center is calculated, and it is divided into the cluster level with the closest distance. The cluster center and the number of clusters come from the clustering result of step S1330.

[0133] For example, it is assumed that the vehicle computers are divided into three screen hardware levels through step S1330: low configuration, medium configuration, and high configuration, and their cluster centers are: Low configuration: ; The cluster center for low-end screen hardware level; Middle configuration: ; It is the cluster center of the mid-range screen hardware level; High configuration: ; The cluster center is the high-end screen hardware level; Hardware parameter vector of a newly launched SUV's in-vehicle screen for: , calculate its distance from the three cluster centers:

[0134] The smallest distance is , so the target SUV is classified as mid-range.

[0135] According to the screen hardware level of the target vehicle computer, the corresponding function classification scheme and personalized function recommendation list are retrieved. The function classification scheme comes from step S2000, and differentiated function classifications are preset for different screen hardware levels. The personalized function recommendation list is generated by comprehensively considering the hardware level, target vehicle model positioning, user preferences, etc. According to the retrieved function classification scheme, all functions supported by the vehicle model are sorted out to form an initial function list. Functions that do not match the positioning of the vehicle model and have low user acceptance are eliminated to obtain a streamlined list of candidate vehicle computer functions. Functions suitable for the vehicle computer screen hardware level are screened out from the candidate vehicle computer function list. For different levels of vehicle computers, the form and complexity of the functions will vary: Low-end car machine: Choose the basic version, with a simple interface layout, mainly buttons and touch for interaction, and functions that focus on ensuring usability.

[0136] Mid-range car computer: Choose the advanced version, the interface is finely designed, and voice and gesture interaction are added. The functions are both practical and beautiful.

[0137] High-end car machine: Choose the flagship version, which has rich interface effects, advanced interactive methods, and integrates technologies such as AR and smart recommendations. The functions emphasize luxury and technology.

[0138] Add the items in the personalized function recommendation list to the filtered function list. The personalized function recommendation list takes into account factors such as the user portrait of the target model, and can improve the user experience and loyalty of the model. Sort the final function list, highlight the key functions, and optimize the layout to generate a set of function solutions that match the target vehicle screen hardware. This solution fully considers the characteristics of the vehicle hardware in terms of function content and display form.

[0139] Step S3000 screens and supplements vehicle functions at different granularities, seeks matching in multiple dimensions such as hardware level, vehicle model positioning, and user preferences, maximizes the use of hardware performance, and takes into account user experience and product features. Compared with simply stacking functions, hierarchical adaptation can achieve reasonable configuration of functions within limited hardware resources. The overall optimization and arrangement of the function list not only highlights the display of key functions, but also takes into account the aesthetic coordination of the layout, which can enhance the user experience at both the functional and visual levels.

[0140] For example, the hardware parameters of a mid-range sedan's car screen are: screen size 8 inches, resolution 1280*720, pixel density 210ppi, color gamut coverage 95% sRGB, CPU 1.8GHz, memory 4GB, storage 32GB. Follow the above steps: 1. After preprocessing the vehicle screen hardware parameters, a 7-dimensional screen hardware parameter vector (210, 95%, 1.8, 600, 4, 32, 8) is obtained, where 600 is the estimated GPU frequency.

[0141] 2. Calculate the distance between the vector and the cluster center. If it is closest to the mid-range cluster center, it is classified into the mid-range hardware level.

[0142] 3. Retrieve the function classification scheme designed for mid-range hardware, as well as a personalized function recommendation list for users of this model.

[0143] 4. Filter the navigation, music, Bluetooth phone and other functions supported by the car, eliminate high-end functions such as mobile phone screen projection, AR navigation, etc., and retain advanced functions with beautiful interface and friendly interaction.

[0144] 5. Add special features such as Internet of Vehicles services and driving behavior analysis from the personalized feature recommendation list.

[0145] 6. Sort and layout the personalized function recommendation list, highlighting the main functions such as music and navigation, and arrange other functions according to frequency of use and logical relationships.

[0146] Finally, a list of vehicle functions that not only matches the vehicle hardware performance, but also reflects the vehicle model characteristics and takes into account user preferences is output. This list is graded and differentiated, and the limited hardware resources are reasonably allocated to various functions, which can smoothly realize functional requirements and bring a high-quality human-vehicle interaction experience.

[0147] The solution of this embodiment can tailor a set of matching function lists for different vehicle models based on the hardware heterogeneity, and achieve adaptation at multiple levels of hardware, users, and products, which has strong practical value. This hierarchical adaptation method can be extended to the development of more vehicle-mounted systems.

[0148] Example 2 This embodiment provides a cross-platform vehicle function classification and adaptation optimization system based on embodiment 1, such as Figure 8 As shown, including: Vehicle computer classification module: obtains screen hardware parameters of different vehicle models, maps the screen hardware parameters to the same vector space for clustering, and divides the vehicle computers into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Personalized recommendation module: Design a corresponding vehicle function classification scheme for each screen hardware level; collect different users' usage behavior data on vehicle functions to generate user preference profiles; combine the user preference profiles and the vehicle function classification scheme to generate a personalized function recommendation list; Adaptation module: For the target vehicle computer to be adapted, obtain the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determine the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieve the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generate a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer.

[0149] In the vehicle classification module, the acquisition of screen hardware parameters of different vehicle models includes: Step S1100, obtaining vehicle screen parameters of different vehicle models; the vehicle screen parameters include vehicle screen size parameters, vehicle screen resolution parameters, vehicle screen pixel density parameters and vehicle screen color gamut parameters; Step S1200, obtaining the hardware configuration parameters of the vehicle system of each vehicle model; the hardware configuration parameters include CPU model, memory capacity parameters, GPU model and storage device parameters; In the vehicle computer classification module, mapping the screen hardware parameters to the same vector space for clustering includes: Step S1310, preprocessing and feature selection are performed on screen hardware parameters to obtain cluster features; Step S1320, using the clustering features obtained in step S1310 as the coordinate axes of the multidimensional vector space, and representing the screen hardware parameters of each vehicle computer as an n1-dimensional screen hardware parameter vector in the multidimensional vector space; Step S1330, taking the screen hardware parameter vector of the vehicle computer as input, using the K-means clustering algorithm to divide the vehicle computer systems of different models into n2 screen hardware levels.

[0150] The step S1310 includes: Step S1311, cleaning the screen hardware parameters to remove obvious abnormal values ​​and missing values; Step S1312, normalizing the hardware parameters of the cleaned screen; Step S1313, removing redundant parameters from the normalized screen hardware parameters; constructing parameter pairs according to the normalized screen hardware parameters, calculating the Pearson correlation coefficient between the parameter pairs, and for parameter pairs whose absolute values ​​of the correlation coefficients are greater than a preset correlation threshold, removing one parameter in the parameter pair and leaving the other parameter as a retained parameter; Step S1314, performing principal component analysis on the screen hardware parameters after removing redundant parameters, calculating the variance contribution rate of each principal component, and taking the principal component whose variance contribution rate exceeds a preset contribution rate threshold as a clustering feature.

[0151] In the personalized recommendation module, the corresponding vehicle function classification scheme is designed for each screen hardware level, including: Step S2100, designing a corresponding vehicle machine function classification scheme for each screen hardware level; the vehicle machine function classification scheme includes three levels: basic functions, extended functions, and advanced functions; Step S2200, collecting usage behavior data of different users on vehicle computer functions to generate user preference profiles; Step S2300, for each screen hardware level, combining the user preference portrait and the vehicle machine function classification scheme, generates a personalized function recommendation list.

[0152] The step S2100 includes: Step S2110, obtaining computing performance indexes of vehicle computers with different screen hardware levels according to hardware configuration parameters of vehicle computers with different screen hardware levels; Step S2120, obtaining display effect indexes of vehicle computers with different screen hardware levels according to vehicle computer screen parameters of vehicle computers with different screen hardware levels; Step S2130, for each screen hardware level, the vehicle computer functions are divided into three levels: basic functions, extended functions, and advanced functions according to the computing performance index and display effect index of the vehicle computer, and combined with the functional complexity and resource occupancy level.

[0153] The step S2130 includes: Step S2131, for each screen hardware level, calculating the function complexity index of the vehicle computer; Step S2132, for each screen hardware level, calculating the resource occupation index of the vehicle computer; Step S2133, for each screen hardware level, according to the computing performance index, display effect index, functional complexity index and resource occupancy index of the vehicle computer, the support vector machine (SVM) algorithm is used to construct a function classification model, and the vehicle computer functions are divided into three levels: basic functions, extended functions and advanced functions.

[0154] The step S2200 includes: Step S2210, collecting usage behavior data of different users on vehicle computer functions; the usage behavior data includes user demographic characteristics and function usage records; Step S2220, constructing the demographic characteristics and function usage records of each user into a user behavior feature vector, clustering the user behavior feature vector, and dividing the users into n3 user groups; Step S2230, statistics the average usage frequency and duration of different functions by each user group, construct a user-function preference matrix, and generate a function preference portrait of each user group; the rows of the user-function preference matrix represent different user groups, the columns represent various vehicle functions, and the matrix elements represent the degree of preference of a certain user group for a certain vehicle function.

[0155] The step S2300 includes: Step S2310, based on the user preference portrait, obtaining a list of preferred functions of different user groups; Step S2320, for each screen hardware level, filter the high-frequency preference functions from the corresponding vehicle machine function classification scheme to obtain a candidate function set; Step S2330: Generate a personalized function recommendation list from the candidate function set.

[0156] The step S2310 includes: Step S2311, sorting each row of the user-function preference matrix in descending order to obtain a function ranking list of each user group's preference from high to low; Step S2312, intercepting the N preferred functions with the highest preference degree in each function ranking list, and generating a preferred function list for each user group; Step S2313 , counting the occurrence frequency of each preference function in the preference function list, marking the preference functions whose occurrence frequency exceeds a preset frequency threshold as high-frequency preference functions, and generating a high-frequency preference function set.

[0157] The step S2320 includes: Step S2321, including all basic functions and extended functions in the high-frequency preference function set into the candidate function set; Step S2322, calculating the resource occupation level SE of each high-level function in the high-frequency preference function set; Step S2323: The advanced functions in the high-frequency preference function set whose resource occupation level SE is lower than a preset resource occupation level threshold are included in the candidate function set.

[0158] The step S2330 includes: Step S2331, obtaining demographic characteristics and historical function usage records of the user to be recommended, constructing a user behavior feature vector of the user to be recommended; calculating the similarity between the user behavior feature vector of the user to be recommended and the user behavior feature vectors of each user group, and finding the user group with the highest similarity as the user group to which the user to be recommended belongs; Step S2332, obtaining a list of preferred functions of the user group to which the user to be recommended belongs, and finding the intersection with the candidate function set to obtain a candidate recommended function set; Step S2333: Based on the historical function usage record of the user to be recommended, the historical function preferences of the user to be recommended are obtained, and according to the historical function preferences of the user to be recommended, M functions are selected from the candidate recommended function set to generate a personalized function recommendation list.

[0159] Example 3 This embodiment discloses an electronic device, which may include one or more processors and one or more memories. The memories store computer-readable codes, which, when executed by the one or more processors, may execute the above-mentioned cross-platform vehicle machine function classification and adaptation optimization method.

[0160] The method or system according to the implementation of the present application can also be implemented with the help of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. A storage device in an electronic device, such as a ROM or a hard disk, can store a cross-platform vehicle function classification and adaptation optimization method provided in the present application. A cross-platform vehicle computer function classification and adaptation optimization method may, for example, include: obtaining screen hardware parameters of different vehicle models, mapping the screen hardware parameters to the same vector space for clustering, and dividing the vehicle computer into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; for each screen hardware level, designing a corresponding vehicle computer function classification scheme; collecting different users' usage behavior data on vehicle computer functions to generate user preference portraits; combining the user preference portraits and the vehicle computer function classification scheme to generate a personalized function recommendation list; for the target vehicle computer to be adapted, obtaining the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determining the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieving the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generating a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer.

[0161] Furthermore, the electronic device may also include a user interface. Of course, the architecture disclosed in the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0162] Example 4 This embodiment discloses a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, a cross-platform vehicle machine function classification and adaptation optimization method of an embodiment of the present application can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0163] In addition, according to the implementation mode of the present application, the process described with reference to the flowchart above can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, the non-temporary machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: obtaining screen hardware parameters of different vehicle models, mapping the screen hardware parameters to the same vector space for clustering, and dividing the vehicle computer into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; for each screen hardware level, design a corresponding vehicle computer function classification scheme; collect different users' usage behavior data on vehicle computer functions, and generate user preference portraits; combine the user preference portraits and the vehicle computer function classification scheme to generate a personalized function recommendation list; for the target vehicle computer to be adapted, obtain the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determine the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieve the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generate a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are performed.

[0164] The methods, systems, and devices of the present application may be implemented in many ways. For example, the methods, systems, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0165] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0166] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cross-platform vehicle machine function classification and adaptation optimization method, characterized in that: The method comprises: Obtain screen hardware parameters of different vehicle models, map the screen hardware parameters to the same vector space for clustering, and divide the vehicle computer into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Design a corresponding vehicle function classification scheme for each screen hardware level; collect different users' usage behavior data on vehicle functions to generate user preference profiles; combine the user preference profiles and the vehicle function classification scheme to generate a personalized function recommendation list; For the target vehicle computer to be adapted, the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer are obtained, the screen hardware level of the target vehicle computer is determined, and according to the screen hardware level of the target vehicle computer, the corresponding vehicle computer function classification scheme and personalized function recommendation list are retrieved, and a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer is adaptively generated.

2. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 1 is characterized in that: Mapping the screen hardware parameters to the same vector space for clustering includes: Preprocess and select features of screen hardware parameters to obtain clustering features; The clustering features are used as coordinate axes of a multidimensional vector space, and the screen hardware parameters of each vehicle computer are represented as an n1-dimensional screen hardware parameter vector in the multidimensional vector space; Taking the screen hardware parameter vector of the vehicle computer as input, the K-means clustering algorithm is used to divide the vehicle computer systems of different models into n2 screen hardware levels.

3. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 2 is characterized in that: The clustering features obtained by preprocessing and feature selection of the screen hardware parameters include: Clean the screen hardware parameters to remove outliers and missing values; Normalize the hardware parameters of the screen after cleaning; Eliminate redundant parameters of the normalized screen hardware parameters; construct parameter pairs based on the normalized screen hardware parameters, calculate the Pearson correlation coefficient between the parameter pairs, and for parameter pairs whose absolute value of the correlation coefficient is greater than a preset correlation threshold, eliminate one parameter in the parameter pair and keep the other parameter as the retained parameter; The principal component analysis is performed on the screen hardware parameters after the redundant parameters are eliminated, the variance contribution rate of each principal component is calculated, and the principal component whose variance contribution rate exceeds the preset contribution rate threshold is used as the clustering feature.

4. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 1 is characterized in that: The design of the corresponding vehicle function classification scheme for each screen hardware level includes: According to the hardware configuration parameters of the vehicle computer, the computing performance index of the vehicle computer is obtained; According to the vehicle computer screen parameters, the display effect index of the vehicle computer is obtained; Calculate the functional complexity index and resource occupancy index of the vehicle computer; According to the computing performance index, display effect index, functional complexity index and resource occupancy index of the vehicle computer, the support vector machine (SVM) algorithm is used to construct a functional classification model, which divides the vehicle computer functions into three levels: basic functions, extended functions and advanced functions.

5. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 1 is characterized in that: The usage behavior data includes the user's demographic characteristics and function usage records; Generating a user preference profile includes: The demographic characteristics and function usage records of each user are constructed into a user behavior feature vector, and the user behavior feature vector is clustered to divide the users into n3 user groups; The average usage frequency and duration of different functions by each user group are counted, a user-function preference matrix is ​​constructed, and a function preference portrait of each user group is generated; the rows of the user-function preference matrix represent different user groups, the columns represent various vehicle computer functions, and the matrix elements represent the degree of preference of a certain user group for a certain vehicle computer function.

6. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 5 is characterized in that: Generating a personalized function recommendation list includes: Based on user preference profiles, obtain a list of preferred functions for different user groups; For each screen hardware level, filter the high-frequency preferred functions from the corresponding vehicle machine function classification scheme to obtain a candidate function set; Generate a personalized feature recommendation list from the candidate feature set.

7. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 6 is characterized in that: The obtaining of the preference function lists of different user groups includes: Sort each row of the user-function preference matrix in descending order to obtain a function ranking list from high to low preference for each user group; Intercept the N preferred functions with the highest preference in each function ranking list to generate a preferred function list for each user group; The occurrence frequency of each preference function in the preference function list is counted, and the preference functions whose occurrence frequency exceeds a preset frequency threshold are marked as high-frequency preference functions to generate a high-frequency preference function set.

8. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 7 is characterized in that: The high-frequency preference functions are screened from the corresponding vehicle function classification scheme to obtain a candidate function set including: Include all basic functions and extended functions in the high-frequency preference function set into the candidate function set; Calculate the resource occupation level SE of each high-level function in the high-frequency preference function set; The high-level functions in the high-frequency preference function set whose resource occupancy level SE is lower than a preset resource occupancy level threshold are included in the candidate function set.

9. The cross-platform vehicle machine function classification and adaptation optimization method according to claim 6 is characterized in that: Generating a personalized function recommendation list from the candidate function set includes: Obtain demographic characteristics and historical function usage records of the user to be recommended, and construct a user behavior feature vector of the user to be recommended; calculate the similarity between the user behavior feature vector of the user to be recommended and the user behavior feature vectors of each user group, and find the user group with the highest similarity as the user group to which the user to be recommended belongs; Obtain the preferred function list of the user group to which the recommended user belongs, and find the intersection with the candidate function set to obtain the candidate recommended function set; Based on the historical function usage records of the user to be recommended, the historical function preferences of the user to be recommended are obtained, and according to the historical function preferences of the user to be recommended, M functions are selected from the candidate recommendation function set to generate a personalized function recommendation list.

10. A cross-platform vehicle-machine function classification and adaptation optimization system, which is used to implement a cross-platform vehicle-machine function classification and adaptation optimization method according to any one of claims 1 to 9, characterized in that: The system comprises: Vehicle computer classification module: obtains screen hardware parameters of different vehicle models, maps the screen hardware parameters to the same vector space for clustering, and divides the vehicle computers into n2 discrete screen hardware levels; the screen hardware parameters include vehicle computer screen parameters and hardware configuration parameters; Personalized recommendation module: Design a corresponding vehicle function classification scheme for each screen hardware level; collect different users' usage behavior data on vehicle functions to generate user preference profiles; combine the user preference profiles and the vehicle function classification scheme to generate a personalized function recommendation list; Adaptation module: For the target vehicle computer to be adapted, obtain the vehicle computer screen parameters and hardware configuration parameters of the target vehicle computer, determine the screen hardware level of the target vehicle computer, and according to the screen hardware level of the target vehicle computer, retrieve the corresponding vehicle computer function classification scheme and personalized function recommendation list, and adaptively generate a vehicle computer function list that meets the characteristics of the corresponding vehicle model of the target vehicle computer.

Images