Page display method, system and product

Abstract

The embodiment of the invention provides a page display method, system and product, and the method comprises the steps: carrying out the preprocessing of collected original echo waveform data, and obtaining a target time domain signal sequence; obtaining user behavior data of interaction between a user and an operation system and vehicle interior and exterior scene data in a corresponding time period; preprocessing the obtained user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data; performing feature extraction fusion on the target behavior data and the target scene data to obtain scene behavior features; inputting the scene behavior characteristics into a behavior prediction model which is qualified in training to perform user behavior prediction, and obtaining a corresponding prediction result; and executing corresponding operation according to the prediction result. The method aims at pre-executing predicted operation which a user wants to execute based on the current behavior of the user, so that the use experience of the user is improved.

Description

Technical Field

[0001] The present application relates to the field of page display technology, and in particular to a page display method, system and product. Background Art

[0002] With the rapid development of software and mobile Internet, for a better user experience, software-defined cars have become a consensus among various car manufacturers. The complexity of software increases exponentially with the addition of functions. Different from tablets and PCs, in-vehicle systems are more complex. While enriching user functions, this also increases the issue of driving safety. Complex interactions, excessive attention to the screen, and car system interfaces that require more attention and thinking are not our original intention of moving entertainment information systems into cars.

[0003] In the driving environment, users' use of the operating system is discrete, and is more for finding and controlling. However, existing car systems usually lack the ability to fully learn and predict user behavior, especially in multiple pages and complex operation sequences. Summary of the invention

[0004] In view of this, the present application provides a page display method, system and product, which are intended to pre-execute the operations that the user wants to perform based on the user's current behavior, thereby improving the user experience.

[0005] A first aspect of the present application provides a page display method, the method comprising:

[0006] Preprocess the collected original echo waveform data to obtain the target time domain signal sequence;

[0007] Obtain user behavior data of the user's interaction with the operating system and vehicle and in-vehicle scene data of the corresponding time period;

[0008] Preprocessing the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data;

[0009] Performing feature extraction and fusion on the target behavior data and the target scene data to obtain scene behavior features;

[0010] Inputting the scenario behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results;

[0011] According to the prediction result, corresponding operations are performed.

[0012] Optionally, performing feature extraction and fusion on the target behavior data and the target scene data to obtain scene behavior features includes:

[0013] Performing time alignment processing on the target behavior data and the target scene data;

[0014] The target behavior data after time alignment is subjected to behavior sequence features of the corresponding time period, and association rules between the behavior sequence features and the final operation are extracted to obtain corresponding target behavior features;

[0015] Extracting the human features, the vehicle interior and exterior environment features, and the geographic location features from the target scene data after time alignment to obtain the corresponding target scene features;

[0016] The target behavior feature and the target scene feature are merged by splicing to obtain an initial scene behavior feature;

[0017] When the data amount of the initial scene behavior feature meets the first condition, reducing the dimension of the initial scene behavior feature to obtain the scene behavior feature;

[0018] When the data amount of the initial scene behavior feature does not satisfy the first condition, the initial scene behavior feature is determined to be a scene behavior feature.

[0019] Optionally, when the data volume of the initial scene behavior feature satisfies the first condition, reducing the dimension of the initial scene behavior feature to obtain the scene behavior feature includes:

[0020] In the case where the data amount of the initial scene behavior feature meets the first condition, determining the principal component features in the initial scene behavior feature by principal component analysis;

[0021] The initial scene behavior features are converted to a new feature space through the principal component features to perform dimensionality reduction, thereby obtaining corresponding scene behavior features.

[0022] Optionally, performing corresponding operations according to the prediction result includes:

[0023] Determining a confidence level corresponding to the prediction result;

[0024] When the confidence level satisfies the second condition, executing an operation corresponding to the prediction result;

[0025] When the confidence level does not satisfy the second condition, continue to obtain new user behavior data of the user interacting with the operating system and vehicle interior and exterior scene data of the corresponding time period;

[0026] A new round of user behavior prediction is performed based on the acquired new user behavior data and the in-vehicle and out-of-vehicle scene data of the corresponding time period, as well as the scene behavior features used in the previous round of prediction.

[0027] Optionally, preprocessing the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data includes:

[0028] Cleaning the acquired user behavior data and the acquired in-vehicle and out-vehicle scene data for invalid data to obtain first user behavior data and first in-vehicle and out-vehicle scene data;

[0029] Perform random noise removal on the first user behavior data and the first vehicle interior and exterior scene data to obtain second user behavior data and second vehicle interior and exterior scene data;

[0030] Filling missing data on the second user behavior data and the second vehicle interior and exterior scene data to obtain third user behavior data and third vehicle interior and exterior scene data;

[0031] The third user behavior data and the third in-vehicle and out-of-vehicle scene data are normalized to obtain target behavior data and target scene data.

[0032] Optionally, the method further includes:

[0033] Determine the user's operation behavior at different time periods based on a large number of historical prediction results;

[0034] According to the user's operation behavior in different time periods, the page layout is dynamically adjusted to a target page layout that matches the operation behavior in the corresponding time period.

[0035] Optionally, the method further includes:

[0036] Determine the size of the corresponding touch area according to the interactive operation performed by the user on the screen;

[0037] According to the touch area size, the icons in the page are adjusted to a target size that matches the touch area size.

[0038] Optionally, after performing a corresponding operation according to the prediction result, the method further includes:

[0039] Obtain user feedback behavior data of the user's interaction with the operating system after performing the corresponding operation;

[0040] According to the user feedback behavior data, the accuracy of the prediction result of the behavior prediction model is evaluated to obtain a corresponding evaluation result;

[0041] Based on the evaluation results, the behavior prediction model is updated and learned online.

[0042] Optionally, the method further includes:

[0043] Monitoring and determining the performance of the behavior prediction model;

[0044] The behavior prediction model is optimized according to the performance.

[0045] A second aspect of the present application provides a page display system, the system comprising:

[0046] A data acquisition module, used to acquire user behavior data of the user's interaction with the operating system and vehicle and vehicle scene data in a corresponding period of time;

[0047] A preprocessing module, used to preprocess the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data;

[0048] A feature fusion module, used to extract and fuse the target behavior data and the target scene data to obtain scene behavior features;

[0049] A behavior prediction module is used to input the scene behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results;

[0050] The control module is used to perform corresponding operations according to the prediction results.

[0051] The third aspect of the present application provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and running on the processor, wherein when the computer program is executed by the processor, the steps in a page display method as described in the first aspect of the present application are implemented.

[0052] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the page display method described in the first aspect of the present application are implemented.

[0053] The page display method provided by this application has the following advantages:

[0054] A page display method provided in an embodiment of the present application first obtains user behavior data of the user interacting with the operating system and the scene data inside and outside the vehicle in a corresponding period of time; pre-processes the obtained user behavior data and the scene data inside and outside the vehicle to obtain target behavior data and target scene data; extracts and fuses the target behavior data and the target scene data to obtain scene behavior features; inputs the scene behavior features into a trained qualified behavior prediction model to predict user behavior and obtain corresponding prediction results; and performs corresponding operations according to the prediction results. Therefore, the present application uses the user behavior data of the user interacting with the operating system and the scene data inside and outside the vehicle in a corresponding period of time to predict the operation that the user wants to perform (such as the page to which he wants to jump), so that environmental factors are taken into account in the prediction process, and the behavior prediction is more accurate, which can effectively improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0056] Figure 1 A flow chart of a page display method shown in one embodiment of the present application;

[0057] Figure 2 A schematic diagram of a page display system according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0059] refer to Figure 1 , Figure 1 This is a flow chart of a page display method shown in one embodiment of the present application. Figure 1 As shown, the method includes:

[0060] Step S1: Obtain user behavior data of the user interacting with the operating system and vehicle interior and exterior scene data of the corresponding time period.

[0061] In this embodiment, the user behavior data of the user's interaction with the operating system is obtained by calling the API at the operating system level. The user behavior data includes but is not limited to the order and timestamp of page visits; the user's stay time on each page; the operations performed by the user on the page; such as clicks, slides, inputs, etc., the user's input content, including text and voice commands; device status information, such as screen brightness, volume, etc. By obtaining the in-vehicle and out-of-vehicle scene data in the same time period as the user behavior data, the in-vehicle and out-of-vehicle scene data include but are not limited to weather; the mood of the user in the car; the number of people in the car load and which specific people; the outside environment information and geographic location information, etc. The data type that can be obtained by the vehicle-mounted sensor monitoring is directly obtained by the vehicle-mounted sensor monitoring, and the data type that cannot be obtained by the vehicle-mounted sensor receives the data of the data type sent by the remote cloud.

[0062] Step S2: pre-processing the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data.

[0063] In this embodiment, in order to avoid the influence of some operations such as accidental touches or abnormal in-vehicle and out-of-vehicle scene data on the prediction results, the present application pre-processes the user behavior data and in-vehicle and out-of-vehicle scene data obtained in the same time period to filter out abnormal data therein, thereby obtaining the corresponding target behavior data and target scene data. The pre-processing can be performed on data of different types and formats.

[0064] In the present application, step S2 may include: cleaning the acquired user behavior data and the inside-outside-vehicle scene data for invalid data to obtain first user behavior data and first inside-outside-vehicle scene data; removing random noise from the first user behavior data and the first inside-outside-vehicle scene data to obtain second user behavior data and second inside-outside-vehicle scene data; filling missing data for the second user behavior data and the second inside-outside-vehicle scene data to obtain third user behavior data and third inside-outside-vehicle scene data; normalizing the third user behavior data and the third inside-outside-vehicle scene data to obtain target behavior data and target scene data.

[0065] In this embodiment, invalid data cleaning is first performed on the user behavior data and vehicle interior and exterior scene data obtained in the same time period, such as invalid data such as abnormal page access time and missing operation type, so as to obtain the first user behavior data and the first vehicle interior and exterior scene data corresponding to each other after cleaning. Random noise cleaning is continued for the obtained first user behavior data and the first vehicle interior and exterior scene data, and random noise in the data is removed, such as unintentional clicks or slides by the user, so as to obtain the second user behavior data and the second vehicle interior and exterior scene data after random noise cleaning. Missing data filling is continued for the obtained second user behavior data and the second vehicle interior and exterior scene data, such as filling the missing data with the average value of the previous and next data, so as to obtain the third user behavior data and the third vehicle interior and exterior scene data after missing data filling. Finally, the third user behavior data and the third vehicle interior and exterior scene data are normalized to unify the dimensions of the data for subsequent model processing, such as converting the dwell time into seconds and converting the screen brightness into percentage, so as to obtain the normalized target behavior data and target scene data.

[0066] Step S3: extracting and fusing the target behavior data and the target scene data to obtain scene behavior features.

[0067] In this embodiment, after obtaining the pre-processed target behavior data and target scene data through step S2, feature extraction is performed on the target behavior data and the target scene data respectively, and the extracted features are fused to obtain a fused scene behavior feature, which includes not only the user's behavior characteristics but also the behavior characteristics in specific scenarios (such as the behavior characteristics on the way home in rainy weather during the afternoon after get off work period, and the behavior characteristics on the way to work in clear weather during the morning work period).

[0068] In the present application, step S3 may include: performing time alignment processing on the target behavior data and the target scene data;

[0069] The target behavior data after time alignment is subjected to behavior sequence features of corresponding time periods, and association rules between the behavior sequence features and final operations are extracted to obtain corresponding target behavior features; the target scene data after time alignment is subjected to character features, vehicle interior and exterior environment features, and geographic location features are extracted to obtain corresponding target scene features; the target behavior features and the target scene features are fused by splicing to obtain initial scene behavior features; when the data volume of the initial scene behavior features meets a first condition, the initial scene behavior features are reduced in dimension to obtain scene behavior features; when the data volume of the initial scene behavior features does not meet the first condition, the initial scene behavior features are determined to be scene behavior features.

[0070] In this embodiment, in order to improve the accuracy of behavior prediction, the present application considers not only the specific scenario information of the user performing the corresponding interactive behavior, but also the corresponding time information for the input data used to predict the user behavior. Therefore, for the feature extraction and fusion of step S3 of the present application, the present application provides another implementation method, specifically:

[0071] First, the target behavior data and target scene data obtained by step S2 are time-aligned, and then the collection time corresponding to the target behavior data after the time alignment is determined. According to the collection time, it is determined in which period the data is collected, and the period includes but is not limited to the morning working period, the afternoon off-duty period, the lunch break period at noon, the afternoon period on weekends, etc. Then, based on the determined period, the behavior sequence characteristics of the corresponding period and the association rule between the behavior sequence characteristics and the operation to be finally performed are extracted to obtain the corresponding target behavior characteristics. The extracted target behavior characteristics record the period of collecting the target behavior data corresponding to the behavior sequence characteristics, the user's behavior characteristics, and the association relationship characteristics between the user's behavior characteristics and the final operation (such as the user usually opens the calendar after checking the email). At the same time, the character characteristics in the car are extracted from the target scene data after the time alignment (the character characteristics are used to determine the mood of the user in the car, the number of people in the car load and who is in the car), the characteristics of the internal and external environment of the car (such as weather, road conditions, temperature in the car, noise conditions in the car, etc.) and the geographical location characteristics (such as the current geographical location) to obtain the corresponding target scene characteristics. After obtaining the target behavior features and target scene features through feature extraction, in order to be able to reversely obtain the original data so that the model training and optimization process can be traced back to the original data, this application directly splices and fuses the extracted target behavior features and target scene features when performing feature fusion to obtain the corresponding initial scene behavior features. At the same time, efficiently and quickly obtaining the prediction results to load the user's concerned pages faster in advance or to execute the actions that the user wants to perform faster in advance is the key to improving the user experience, so the prediction efficiency is the content that needs to be focused on. Therefore, after the initial scene behavior features are obtained by fusion, in order to improve the model prediction efficiency, when the data volume of the initial scene behavior features is too large, it is subjected to dimensionality reduction processing to improve the model prediction efficiency. Therefore, after obtaining the initial scene behavior features, this application determines whether the data volume of the initial scene behavior features meets the first condition. If the first condition is met, the initial scene behavior features obtained are subjected to dimensionality reduction processing to obtain the scene behavior features after dimensionality reduction processing, wherein the first condition can be that when the data volume of the initial scene behavior features exceeds the set threshold, it is determined that the first condition is met. The set threshold can be set according to the actual application scenario, and is not specifically limited here. When the first condition is not met, it is determined that the amount of data of the obtained initial scene behavior features is low. In this case, no dimensionality reduction processing is required, and the initial scene behavior features are directly determined as scene behavior features.

[0072] In the present application, when the data volume of the initial scene behavior features meets the first condition, the initial scene behavior features are reduced in dimension to obtain scene behavior features, including: when the data volume of the initial scene behavior features meets the first condition, the principal component features in the initial scene behavior features are determined by principal component analysis; and the initial scene behavior features are converted to a new feature space by using the principal component features to perform dimensionality reduction to obtain corresponding scene behavior features.

[0073] In this embodiment, although obtaining prediction results efficiently and quickly to load the pages that the user is interested in faster or to execute the actions that the user wants to perform faster is the key to improving the user experience, prediction accuracy is also a key point that cannot be ignored in improving the user experience. When the data volume of the initial scene behavior features is too large and the dimension is reduced, it will inevitably affect the original expression of the features, thereby affecting the prediction accuracy. Therefore, in order to balance the accuracy and efficiency of prediction, the present application hopes to reduce the dimension while minimizing the impact on the prediction accuracy. The present application finds that the user's behavioral habits are basically unchanged within a certain period of time. Therefore, the user's operating behavior within a certain period of time basically belongs to a linear relationship. The advantage of principal component analysis (PCA Principal Component Analysis) is that it can retain the main information in the data. This advantage can reduce the impact of dimensionality reduction on prediction accuracy through this method. The disadvantage is that it may ignore some important nonlinear relationships. The user's operating behavior within a certain period of time basically belongs to a linear relationship. Therefore, this disadvantage does not affect the prediction accuracy within a certain period of time. Therefore, the present application adopts the principal component analysis method to reduce the dimension of the initial scene behavior features with excessive data volume. Specifically: when the data volume of the initial scene behavior features meets the first condition, first, when the data volume of the initial scene behavior features meets the first condition, the principal component features in the initial scene behavior features are determined by principal component analysis, and then the initial scene behavior features are converted to a new feature space through the principal component features for dimensionality reduction to obtain the corresponding scene behavior features.

[0074] Step S4: Input the scene behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results.

[0075] In this embodiment, after obtaining the fused scene behavior feature through step S3, the scene behavior feature is input into a well-trained behavior prediction model to predict user behavior and obtain a corresponding prediction result, which represents the operation that the user may want to perform under the scene behavior feature, such as opening a specific page, or opening a specific page to perform a specific operation (such as playing music, playing videos, turning on the air conditioner, locking the windows, etc.). Among them, an optional implementation of the behavior prediction model is a long short-term memory network (LSTM), which is trained end-to-end, that is, directly training the entire process from user behavior data to page display prediction, to obtain a well-trained behavior prediction model.

[0076] Step S5: Execute corresponding operations according to the prediction results.

[0077] In this embodiment, based on the prediction result obtained in step S4, the vehicle is controlled to perform corresponding control operations (such as playing music, playing videos, turning on the air conditioner, locking the windows, etc.).

[0078] A page display method provided in an embodiment of the present application first obtains user behavior data of the user interacting with the operating system and the scene data inside and outside the vehicle in a corresponding period of time; pre-processes the obtained user behavior data and the scene data inside and outside the vehicle to obtain target behavior data and target scene data; extracts and fuses the target behavior data and the target scene data to obtain scene behavior features; inputs the scene behavior features into a trained qualified behavior prediction model to predict user behavior and obtain corresponding prediction results; and performs corresponding operations according to the prediction results. Therefore, the present application uses the user behavior data of the user interacting with the operating system and the scene data inside and outside the vehicle in a corresponding period of time to predict the operation that the user wants to perform (such as the page to which he wants to jump), so that environmental factors are taken into account in the prediction process, and the behavior prediction is more accurate, which can effectively improve the user experience.

[0079] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a page display method. In the page display method, step S5 may include: determining the confidence corresponding to the prediction result; if the confidence satisfies the second condition, executing the operation corresponding to the prediction result; if the confidence does not satisfy the second condition, continuing to obtain new user behavior data of the user interacting with the operating system and the in-vehicle and out-of-vehicle scene data of the corresponding time period; performing a new round of user behavior prediction based on the new user behavior data and the in-vehicle and out-of-vehicle scene data of the corresponding time period, as well as the scene behavior features used in the previous round of prediction.

[0080] In this embodiment, the present application is mainly intended to improve the user experience, and an inaccurate prediction result being executed will make the user experience worse. Therefore, after obtaining the prediction result, the present application will further judge the confidence of the prediction result. If its confidence is greater than or equal to the set threshold, it is determined that the confidence meets the second condition. At this time, the operation corresponding to the prediction result is executed, wherein the set threshold can be set according to the actual application scenario and is not specifically limited here. When the confidence level is less than the set threshold, it is determined that the confidence level does not meet the second condition, and the prediction result will not guide the execution of the corresponding operation. At this time, the present application will continue to monitor and obtain new user behavior data of the user interacting with the operating system and the in-vehicle and out-of-vehicle scene data in the same time period as the new user behavior data. For the newly obtained user behavior data and the in-vehicle and out-of-vehicle scene data, the two are fused through the same above-mentioned implementation method to obtain corresponding scene behavior features, and the scene behavior features are further fused with the scene behavior features used in the previous round of prediction to form new scene behavior features to be used for behavior prediction. The scene behavior features are then input into a qualified trained behavior prediction model to perform user behavior prediction to obtain a corresponding new round of prediction results. If the confidence level of the prediction result meets the second condition, the corresponding action is executed. If it still does not meet the second condition, a new round of prediction is continued.

[0081] For example, assuming that the scene behavior feature used in the previous round of prediction is A1, the scene behavior feature A2 is obtained by fusing the newly acquired user behavior data and the scene data inside and outside the car, and A1 and A2 are fused to obtain the new scene behavior feature A1+A2. At this time, the scene behavior feature A1+A2 includes more behavioral actions of the user interacting with the operating system, and the corresponding prediction results will be more accurate. Therefore, when the confidence of the prediction result is insufficient, a sufficiently accurate prediction result can be obtained after multiple rounds of prediction. At the same time, the scene behavior feature A1+A2 currently used for prediction will only be fused with the newly acquired user behavior data and the scene data inside and outside the car corresponding to the scene behavior feature A2, and will no longer be fused with the user behavior data and the scene data inside and outside the car corresponding to the scene behavior feature A1, so the processing efficiency will also be higher.

[0082] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a page display method. In the page display method, the method further includes: determining the user's operation behavior in different time periods based on a large number of historical prediction results; and dynamically adjusting the page layout to a target page layout that matches the operation behavior in the corresponding time period based on the user's operation behavior in different time periods.

[0083] In this embodiment, after obtaining a large number of historical prediction results for the user, the user's operation behavior at different time periods is determined through statistical analysis. Based on the determined user's operation behavior at different time periods, the page layout is dynamically adjusted to the target page layout that matches the operation behavior at the corresponding time period. For example, the user will listen to music during the afternoon off-get off work period every day, and the music type is soothing. Correspondingly, during the afternoon off-get off work period every day, the music player software is adjusted to the first interface displayed on the screen, so that the user can open the music player as soon as the screen is turned on, and put the soothing music in the music player to the first interface of the music player; and the user will lock the rear windows during the morning work period every day, and correspondingly, during the morning work period every day, the window control button is adjusted to the first interface displayed on the screen, so that the user can lock the rear windows as soon as the screen is turned on.

[0084] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a page display method. In the page display method, the method further includes: determining the corresponding touch area size according to the interactive operation performed by the user on the screen; and adjusting the icons in the page to a target size matching the touch area size according to the touch area size.

[0085] In this embodiment, when different users operate the screen, the range of contact between their fingers and the screen is different. For example, when some users perform operations such as clicking or sliding the screen, the contact range is larger, while some users have a smaller contact range. Therefore, in order to prevent users from accidentally touching displayed charts, etc., this application determines the size of the touch area when the user touches the screen based on the interactive operations performed by the user on the screen. Then, based on the determined touch area size, the icons on the page are adjusted to a target size that matches the touch area size to prevent users from accidentally touching. That is, the larger the determined touch area size, the larger the icons on the page are adjusted to.

[0086] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a page display method. In the page display method, after executing the corresponding operation according to the prediction result, the method further includes: obtaining user feedback behavior data of the user interacting with the operating system after executing the corresponding operation; evaluating the accuracy of the prediction result of the behavior prediction model according to the user feedback behavior data to obtain the corresponding evaluation result; and performing online update learning on the behavior prediction model according to the evaluation result.

[0087] In this embodiment, after executing the corresponding operation based on the prediction result, the operation behavior of the user interacting with the operating system is determined, and the operation behavior is determined as user feedback behavior data. Then, the user feedback behavior data is analyzed to evaluate the accuracy of the prediction results of the qualified behavior prediction model, and the corresponding evaluation results are obtained. When the evaluation results characterize that the prediction results of the behavior prediction model are not accurate enough, new user behavior data is collected again to further update and learn the behavior prediction model online. For example, after jumping to the music player and starting soothing music playback, it is determined that the operation behavior of the user interacting with the operating system is to exit the music player, and it is determined that the prediction results of the behavior prediction model are not accurate enough. At this time, the qualified behavior prediction model is further updated and learned online.

[0088] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a page display method. In the page display method, the method further includes: monitoring and determining the performance of the behavior prediction model; and optimizing the behavior prediction model according to the performance.

[0089] In this embodiment, not only the accuracy of the behavior prediction model directly affects the user experience, but its performance also directly affects the user experience. For example, if the behavior prediction model takes a long time or the user performs many steps of behavior operations to get an accurate prediction result, the user experience is also poor. Therefore, this application will monitor and determine the performance of the behavior prediction model, which includes but is not limited to accuracy, prediction efficiency, etc. When its performance is insufficient, the behavior prediction model is updated and learned online.

[0090] Based on the same inventive concept, an embodiment of the present application provides a page display system, such as Figure 2 As shown, the system 200 includes:

[0091] The data acquisition module 201 is used to acquire the user behavior data of the user interacting with the operating system and the in-vehicle and out-of-vehicle scene data of the corresponding time period;

[0092] A preprocessing module 202 is used to preprocess the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data;

[0093] A feature fusion module 203 is used to extract and fuse the target behavior data and the target scene data to obtain scene behavior features;

[0094] The behavior prediction module 204 is used to input the scene behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results;

[0095] The control module 205 is used to perform corresponding operations according to the prediction results.

[0096] Optionally, the feature fusion module 203 includes:

[0097] An alignment processing module, used for performing time alignment processing on the target behavior data and the target scene data;

[0098] A first feature extraction module is used to extract the behavior sequence features of the corresponding time period from the target behavior data after the time alignment process, and to extract the association rules between the behavior sequence features and the final operation, so as to obtain the corresponding target behavior features;

[0099] The second feature extraction module is used to extract the human features, the vehicle interior and exterior environment features and the geographical location features of the target scene data after the time alignment process to obtain the corresponding target scene features;

[0100] A first feature fusion module, used to fuse the target behavior feature and the target scene feature by splicing to obtain an initial scene behavior feature;

[0101] A second feature fusion module is used to reduce the dimension of the initial scene behavior feature to obtain the scene behavior feature when the data amount of the initial scene behavior feature meets the first condition;

[0102] The third feature fusion module is used to determine that the initial scene behavior feature is a scene behavior feature when the data amount of the initial scene behavior feature does not meet the first condition.

[0103] Optionally, the second feature fusion module includes:

[0104] A feature determination module, configured to determine a principal component feature in the initial scene behavior feature by principal component analysis if the data amount of the initial scene behavior feature meets a first condition;

[0105] The second feature fusion submodule is used to convert the initial scene behavior features into a new feature space through the principal component features to perform dimensionality reduction and obtain corresponding scene behavior features.

[0106] Optionally, the control module 205 includes:

[0107] A confidence determination module, used to determine the confidence corresponding to the prediction result;

[0108] A first control module, configured to execute an operation corresponding to the prediction result when the confidence level satisfies a second condition;

[0109] A data acquisition module, configured to continue acquiring new user behavior data of the user interacting with the operating system and vehicle interior and exterior scene data of a corresponding period of time when the confidence level does not satisfy the second condition;

[0110] The behavior prediction submodule is used to perform a new round of user behavior prediction based on the acquired new user behavior data and the in-vehicle and out-of-vehicle scene data of the corresponding time period, as well as the scene behavior features used in the previous round of prediction.

[0111] Optional, pre-processing modules, including:

[0112] A data cleaning module, used to clean invalid data from the acquired user behavior data and the acquired in-vehicle and out-vehicle scene data, to obtain first user behavior data and first in-vehicle and out-vehicle scene data;

[0113] A noise removal module, configured to perform random noise removal on the first user behavior data and the first vehicle interior and exterior scene data to obtain second user behavior data and second vehicle interior and exterior scene data;

[0114] A data filling module, used to fill missing data in the second user behavior data and the second vehicle interior and exterior scene data to obtain third user behavior data and third vehicle interior and exterior scene data;

[0115] The normalization processing module is used to normalize the third user behavior data and the third in-vehicle and out-of-vehicle scene data to obtain target behavior data and target scene data.

[0116] Optionally, the system 200 further includes:

[0117] The operation behavior determination module is used to determine the user's operation behavior in different time periods based on a large number of historical prediction results;

[0118] The page layout adjustment module is used to dynamically adjust the page layout to a target page layout that matches the operation behavior in the corresponding time period according to the user's operation behavior in different time periods.

[0119] Optionally, the system 200 further includes:

[0120] A size determination module, used to determine the corresponding touch area size according to the interactive operation performed by the user on the screen;

[0121] The size adjustment module is used to adjust the icons in the page to a target size matching the touch area size according to the touch area size.

[0122] Optionally, the system 200 further includes:

[0123] A data acquisition module, used to acquire user feedback behavior data of the user's interaction with the operating system after performing a corresponding operation;

[0124] An evaluation module, used to evaluate the accuracy of the prediction results of the behavior prediction model according to the user feedback behavior data, and obtain a corresponding evaluation result;

[0125] A model updating module is used to perform online updating and learning of the behavior prediction model according to the evaluation result.

[0126] Optionally, the system 200 further includes:

[0127] A monitoring module, used to monitor and determine the performance of the behavior prediction model;

[0128] An optimization module is used to optimize the behavior prediction model according to the performance.

[0129] Based on the same inventive concept, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and running on the processor. When the computer program is executed by the processor, the steps in a page display method described in the first aspect of the present application are implemented.

[0130] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in a page display method described in the first aspect of the present application are implemented.

[0131] As for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0132] It should be noted that, for the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0133] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0134] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0135] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0138] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0139] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0140] The above is a detailed introduction to a page display method, system and product provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A page display method, characterized in that: The method comprises: Obtain user behavior data of the user's interaction with the operating system and vehicle and in-vehicle scene data of the corresponding time period; Preprocessing the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data; Performing feature extraction and fusion on the target behavior data and the target scene data to obtain scene behavior features; Inputting the scenario behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results; According to the prediction result, corresponding operations are performed.

2. A page display method according to claim 1, characterized in that: Performing feature extraction and fusion on the target behavior data and the target scene data to obtain scene behavior features includes: Performing time alignment processing on the target behavior data and the target scene data; The target behavior data after time alignment is subjected to behavior sequence features of the corresponding time period, and association rules between the behavior sequence features and the final operation are extracted to obtain corresponding target behavior features; Extracting the human features, the vehicle interior and exterior environment features, and the geographic location features from the target scene data after time alignment to obtain the corresponding target scene features; The target behavior feature and the target scene feature are merged by splicing to obtain an initial scene behavior feature; When the data amount of the initial scene behavior feature meets the first condition, reducing the dimension of the initial scene behavior feature to obtain the scene behavior feature; When the data amount of the initial scene behavior feature does not satisfy the first condition, the initial scene behavior feature is determined to be a scene behavior feature.

3. A page display method according to claim 2, characterized in that: When the data amount of the initial scene behavior feature satisfies the first condition, reducing the dimension of the initial scene behavior feature to obtain the scene behavior feature includes: In the case where the data amount of the initial scene behavior feature meets the first condition, determining the principal component features in the initial scene behavior feature by principal component analysis; The initial scene behavior features are converted to a new feature space through the principal component features to perform dimensionality reduction, thereby obtaining corresponding scene behavior features.

4. A page display method according to claim 1, characterized in that: According to the prediction result, corresponding operations are performed, including: Determining a confidence level corresponding to the prediction result; When the confidence level satisfies the second condition, executing an operation corresponding to the prediction result; When the confidence level does not satisfy the second condition, continue to obtain new user behavior data of the user interacting with the operating system and vehicle interior and exterior scene data of the corresponding time period; A new round of user behavior prediction is performed based on the acquired new user behavior data and the in-vehicle and out-of-vehicle scene data of the corresponding time period, as well as the scene behavior features used in the previous round of prediction.

5. A page display method according to claim 1, characterized in that: Preprocessing the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data includes: Cleaning the acquired user behavior data and the acquired in-vehicle and out-vehicle scene data for invalid data to obtain first user behavior data and first in-vehicle and out-vehicle scene data; Perform random noise removal on the first user behavior data and the first vehicle interior and exterior scene data to obtain second user behavior data and second vehicle interior and exterior scene data; Filling missing data on the second user behavior data and the second vehicle interior and exterior scene data to obtain third user behavior data and third vehicle interior and exterior scene data; The third user behavior data and the third in-vehicle and out-of-vehicle scene data are normalized to obtain target behavior data and target scene data.

6. A page display method according to claim 1, characterized in that: The method further comprises: Determine the user's operation behavior at different time periods based on a large number of historical prediction results; According to the user's operation behavior in different time periods, the page layout is dynamically adjusted to a target page layout that matches the operation behavior in the corresponding time period.

7. A page display method according to claim 1, characterized in that: The method further comprises: Determine the size of the corresponding touch area according to the interactive operation performed by the user on the screen; According to the touch area size, the icons in the page are adjusted to a target size that matches the touch area size.

8. A page display method according to claim 1, characterized in that: After performing corresponding operations according to the prediction result, the method further includes: Obtain user feedback behavior data of the user's interaction with the operating system after performing the corresponding operation; According to the user feedback behavior data, the accuracy of the prediction result of the behavior prediction model is evaluated to obtain a corresponding evaluation result; Based on the evaluation results, the behavior prediction model is updated and learned online.

9. A page display method according to claim 1, characterized in that: The method further comprises: Monitoring and determining the performance of the behavior prediction model; The behavior prediction model is optimized according to the performance.

10. A page display system, characterized in that: The system comprises: A data acquisition module, used to acquire user behavior data of the user's interaction with the operating system and vehicle and vehicle scene data in a corresponding period of time; A preprocessing module, used to preprocess the acquired user behavior data and the vehicle interior and exterior scene data to obtain target behavior data and target scene data; A feature fusion module, used to extract and fuse the target behavior data and the target scene data to obtain scene behavior features; A behavior prediction module is used to input the scene behavior features into a trained behavior prediction model to predict user behavior and obtain corresponding prediction results; The control module is used to perform corresponding operations according to the prediction results.

11. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and running on the processor, wherein the computer program, when executed by the processor, implements the steps in a page display method as claimed in claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the page display method according to claims 1 to 9 are implemented.

Images