Intelligent interface optimization method and system based on user behavior analysis

Abstract

The invention provides an intelligent interface optimization method and system based on user behavior analysis, and the method comprises the steps: obtaining user operation data which comprises shortcut key combination, mouse click, dragging operation and corresponding timestamps and coordinate position information; clustering analysis is carried out on the user operation data to obtain different user groups, and the clustering analysis is carried out according to shortcut key use frequency, mouse click frequency and dragging operation duration; for each user group, an operation feature vector is extracted, and the operation feature vector comprises shortcut key combination frequency, mouse click position distribution and dragging operation duration distribution; constructing a user feature matrix according to the operation feature vector; and obtaining an output result of the behavior prediction model.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an intelligent interface optimization method and system based on user behavior analysis. Background Art

[0002] Problem background:

[0003] In the Internet big data image processing system, the user's operation behavior data is very large and complex, including various shortcut key combinations, mouse clicks, dragging and other operations, as well as detailed information such as the timestamps and coordinate positions of these operations. The system needs to collect and process these high-dimensional and high-frequency user operation data in real time, and extract valuable user operation patterns and habit features from them. This requires the system to have efficient data collection, storage and analysis capabilities, be able to quickly respond to user operation requests, and update user portraits and recommendation models in a timely manner.

[0004] However, the operating habits and preferences of different users vary greatly. Some users like to use shortcut keys, while others prefer mouse clicks. How to identify the operating characteristics of different user groups from massive user operation data and provide personalized interface layouts and functional modules for each group is a very challenging problem. The system needs to establish accurate user portraits and behavior prediction models, and dynamically adjust model parameters and recommendation strategies based on users' real-time operation data to ensure the accuracy and real-time nature of the recommendation results. At the same time, the system must also have flexible interface rendering and function combination capabilities, and be able to dynamically generate personalized interface layouts and functional modules based on the preference characteristics of different user groups to provide a smooth interactive experience. Summary of the invention

[0005] The present invention provides an intelligent interface optimization method based on user behavior analysis, which mainly includes:

[0006] Acquire user operation data, which includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information;

[0007] Performing cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations;

[0008] For each of the user groups, extract an operation feature vector, wherein the operation feature vector includes a shortcut key combination frequency, a mouse click position distribution, and a drag operation duration distribution;

[0009] Constructing a user feature matrix according to the operation feature vector;

[0010] A random forest algorithm is used to train a behavior prediction model according to the user feature matrix, wherein the behavior prediction model is used to predict the user's next operation tendency, and the user's next operation tendency includes the probability of using shortcut keys and the mouse click area preference;

[0011] Obtaining an output result of the behavior prediction model;

[0012] The interface layout is dynamically adjusted according to the output result of the behavior prediction model. If the output result of the behavior prediction model shows that the user prefers to use shortcut keys, a shortcut key prompt area is added. If the output result of the behavior prediction model shows that the user prefers mouse operation, the mouse click area layout is optimized.

[0013] The present invention provides an intelligent interface optimization system based on user behavior analysis, which mainly includes:

[0014] A data acquisition module is used to acquire user operation data, wherein the user operation data includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information;

[0015] A cluster analysis module, used to perform cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations;

[0016] A feature extraction module, for extracting an operation feature vector for each user group, wherein the operation feature vector includes a shortcut key combination frequency, a mouse click position distribution, and a drag operation duration distribution;

[0017] A matrix construction module, used to construct a user feature matrix according to the operation feature vector;

[0018] A model training module, for training a behavior prediction model using a random forest algorithm according to the user feature matrix, wherein the behavior prediction model is used to predict the user's next operation tendency, wherein the user's next operation tendency includes a shortcut key usage probability and a mouse click area preference;

[0019] A result acquisition module, used to obtain the output result of the behavior prediction model;

[0020] The interface adjustment module is used to dynamically adjust the interface layout according to the output results of the behavior prediction model. If the output results of the behavior prediction model indicate that the user prefers to use shortcut keys, a shortcut key prompt area is added. If the output results of the behavior prediction model indicate that the user prefers mouse operation, the mouse click area layout is optimized.

[0021] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0022] The present invention discloses an intelligent interface optimization method based on user behavior analysis. By acquiring user operation data, including information such as shortcut keys, mouse clicks and dragging, cluster analysis is performed on the data to identify different user groups. Operation feature vectors are extracted for each group, a user feature matrix is ​​constructed, and a behavior prediction model is trained using a random forest algorithm. The model can predict the user's next operation tendency, such as the probability of using shortcut keys and the preference for mouse click areas. Based on the prediction results, the present invention can dynamically adjust the interface layout, such as adding shortcut key prompts or optimizing the mouse click area, thereby providing a personalized user experience. This method realizes intelligent adaptation of the interface by deeply analyzing user behavior patterns, effectively improving the usability and user satisfaction of the software. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of an intelligent interface optimization method based on user behavior analysis of the present invention.

[0024] Figure 2 A schematic diagram of an intelligent interface optimization method and system based on user behavior analysis of the present invention.

[0025] Figure 3 This is another schematic diagram of an intelligent interface optimization method and system based on user behavior analysis of the present invention.

[0026] Figure 4 It is a structural schematic diagram of an intelligent interface optimization method and system based on user behavior analysis of the present invention. DETAILED DESCRIPTION

[0027] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0028] like Figure 1-4 In this embodiment, a method and system for optimizing an intelligent interface based on user behavior analysis may specifically include:

[0029] Step S101, obtaining user operation data, wherein the user operation data includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information.

[0030] The operation behavior data of users in the process of using the software is obtained, including keyboard shortcut combinations, mouse clicks and drags, and the timestamps and coordinate position information corresponding to these operations. For the obtained user operation behavior data, data preprocessing technology is used to clean and standardize the data, remove noise data, and normalize the timestamp and coordinate position information to obtain standardized user operation behavior data. According to the preset operation behavior pattern library, the standardized user operation behavior data is feature extracted to obtain the feature vector of the user operation behavior, including operation type, operation time, operation location and other features. The clustering algorithm is used to perform cluster analysis on the feature vector of the user operation behavior, and the user operation behavior is divided into different operation modes according to the clustering results, such as shortcut key operation mode, mouse click operation mode, drag operation mode, etc. For different operation modes, corresponding user behavior prediction models are constructed respectively, such as shortcut key operation prediction model based on sequence model, mouse click operation prediction model based on graph model, etc., and the user behavior prediction model under each operation mode is obtained through training. During the software operation, the user's operation behavior data is obtained in real time, and preprocessed and feature extracted to obtain the feature vector of the current user operation. The corresponding user behavior prediction model is selected according to the operation type to predict the user's subsequent operation behavior. According to the user behavior prediction results, the software's user interface and interaction method are dynamically adjusted, such as automatically recommending commonly used shortcut keys according to the user's shortcut key operation habits, and automatically optimizing the button layout according to the user's mouse click position, etc., to improve the user's operation efficiency and experience.

[0031] Step S102, performing cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of the drag operation.

[0032] Obtain the operation data of users in the process of using the software, including indicators such as the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations. Preprocess the obtained user operation data, remove outliers and missing values, and normalize the data to make different indicators comparable. Based on the preprocessed user operation data, use the K-means clustering algorithm to perform cluster analysis on users, and divide users with similar operation behaviors into the same group. Determine the optimal number of clusters through methods such as the elbow rule or silhouette coefficient to obtain different user groups. For each user group, analyze its operation behavior characteristics and summarize the commonalities and personalized needs of the group. According to the characteristics and needs of different user groups, design targeted personalized service solutions, such as providing a list of commonly used shortcut keys, optimizing the mouse click operation process, etc. Implement personalized service solutions in the software. When users use the software, provide corresponding personalized services according to their groups to improve the user experience and efficiency.

[0033] Step S103: extracting an operation feature vector for each user group, wherein the operation feature vector includes shortcut key combination frequency, mouse click position distribution, and drag operation duration distribution.

[0034] Acquire multiple pre-divided user group data, and extract operation feature vectors for each user group. Use statistical analysis methods to calculate the frequency distribution of shortcut key combinations for each user group, and obtain the shortcut key combination frequency feature vector. Use image processing algorithms to perform cluster analysis on the mouse click position coordinate data of each user group, and obtain the mouse click position distribution feature vector. According to the start and end timestamps of the user's drag operation, calculate the drag operation duration data distribution of each user group, and obtain the drag operation duration distribution feature vector. Combine feature vectors such as shortcut key combination frequency, mouse click position distribution, and drag operation duration distribution to form a user group operation feature vector. Use support vector machine algorithm to classify user group operation feature vectors and train user group operation behavior classification model. If the matching degree between the new user's operation feature vector and the operation feature vector of a certain user group exceeds the preset threshold, it is determined that the new user belongs to the corresponding user group.

[0035] Step S104: construct a user feature matrix according to the operation feature vector.

[0036] Obtain the user's operation log data on the application or platform, extract the operation features based on indicators such as operation type, operation frequency, and operation duration, and generate an operation feature vector. For each user, combine the corresponding operation feature vectors to construct the user's feature matrix. Each row of the matrix represents an operation feature vector, and each column represents a feature dimension. Perform dimensionality reduction processing on the user feature matrix, use the principal component analysis (PCA) algorithm to extract the main features of the matrix, reduce the feature dimension, and obtain a compressed user feature matrix. Based on the compressed user feature matrix, use a clustering algorithm (such as K-means) to group users, divide users with similar operation behavior patterns into the same group, and characterize the behavior characteristics of different user groups. In each user group, analyze the user's preference characteristics based on the user's operation type and operation object, build a user preference portrait, and characterize the user's interest preferences in different business scenarios. Based on the behavior patterns and preference characteristics of different user groups, use the association rule mining algorithm to discover the association rules and patterns between the operation behaviors and preference characteristics of different user groups for business decision-making and personalized recommendations. Regularly update user operation log data, recalculate user feature matrix and user group segmentation, dynamically update user portraits, and continuously optimize business strategies and user experience.

[0037] Step S105, using a random forest algorithm, according to the user feature matrix, training a behavior prediction model, the behavior prediction model is used to predict the user's next operation tendency, the user's next operation tendency includes the probability of using shortcut keys and the mouse click area preference.

[0038] Obtain the user's historical operation records, extract user features, and construct a user feature matrix, where the features include the frequency of shortcut key usage, mouse click location distribution, etc.; use the random forest algorithm, take the user feature matrix as input, train the behavior prediction model, and the model output is the probability distribution of the user's next operation tendency; obtain user features based on the user's current operation status, input them into the trained behavior prediction model, and obtain the probability distribution of the user's next operation tendency; from the operation tendency probability distribution, obtain the probability values ​​of shortcut key usage probability and mouse click area preference, and compare them with the preset shortcut key usage threshold and click area threshold respectively; if the shortcut key usage probability is greater than the threshold, it is judged that the user tends to use shortcut keys, and based on the historical shortcut key usage records, predict the shortcut keys that the user is most likely to use; if the mouse click area preference probability is greater than the threshold, it is judged that the user tends to click the mouse, and based on the historical click location distribution, predict the area that the user is most likely to click; return the predicted shortcut keys and click area preferences for interface optimization and interaction guidance to improve user experience.

[0039] Step S106, obtaining the output result of the behavior prediction model.

[0040] Obtain the output result of the behavior prediction model, the output result including the user's behavior intention and user preference; determine the user's current needs according to the user's behavior intention in the output result, and screen out service content matching the user's needs from a preset service content library according to the user's current needs; further screen out personalized service content matching the user's preferences from the service content screened out from the service content library according to the user's preferences in the output result; determine the recommended content recommended to the user according to the personalized service content, and push the recommended content to the user terminal for display; obtain user feedback information on the recommended content, and adjust the parameters of the behavior prediction model according to the user's feedback information, so that the behavior prediction model can more accurately predict the user's behavior intention and preference; when adjusting the parameters of the behavior prediction model, use a gradient descent algorithm to optimize the parameters of the behavior prediction model, and through multiple iterations, continuously improve the prediction accuracy of the behavior prediction model; apply the optimized behavior prediction model to subsequent user behavior prediction and personalized recommendation processes, and continuously improve the accuracy of the recommended content and user experience.

[0041] Step S107, dynamically adjust the interface layout according to the output result of the behavior prediction model. If the output result of the behavior prediction model indicates that the user prefers to use shortcut keys, add a shortcut key prompt area. If the output result of the behavior prediction model indicates that the user prefers mouse operation, optimize the mouse click area layout.

[0042] Obtain the user's historical operation behavior data, including the frequency and mode of using shortcut keys and mouse operations, as training data for the behavior prediction model. Use machine learning algorithms such as decision trees and random forests to establish a user behavior prediction model, and train it based on the user's historical operation behavior data to obtain a trained behavior prediction model. Collect the user's operation behavior data in real time during the user's use of the software, input it into the behavior prediction model, obtain the model's output results, and judge the user's operation preferences. If the model output results show that the user prefers to use shortcut keys, dynamically increase the area of ​​the shortcut key prompt area in the interface, and optimize the layout design of the shortcut keys, such as arranging commonly used shortcut keys in a prominent position. If the model output results show that the user prefers mouse operation, dynamically optimize the layout design of the mouse click area in the interface, such as increasing the button size, optimizing the button position arrangement, etc., to improve the convenience of mouse operation. Continuously track the user's operation behavior, regularly update the behavior prediction model, and dynamically adjust the interface layout according to the latest prediction results output by the model to achieve adaptive optimization of the interface. Evaluate the user experience of the optimized interface through A / B testing and other methods, collect user feedback, continuously improve the behavior prediction model and interface layout optimization strategy, and improve the overall user experience.

[0043] The present invention provides an intelligent interface optimization system based on user behavior analysis, which mainly includes:

[0044] A data acquisition module is used to acquire user operation data, wherein the user operation data includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information;

[0045] A cluster analysis module, used to perform cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations;

[0046] A feature extraction module, for extracting an operation feature vector for each user group, wherein the operation feature vector includes a shortcut key combination frequency, a mouse click position distribution, and a drag operation duration distribution;

[0047] A matrix construction module, used to construct a user feature matrix according to the operation feature vector;

[0048] A model training module, for training a behavior prediction model using a random forest algorithm according to the user feature matrix, wherein the behavior prediction model is used to predict the user's next operation tendency, wherein the user's next operation tendency includes a shortcut key usage probability and a mouse click area preference;

[0049] A result acquisition module, used to obtain the output result of the behavior prediction model;

[0050] The interface adjustment module is used to dynamically adjust the interface layout according to the output results of the behavior prediction model. If the output results of the behavior prediction model indicate that the user prefers to use shortcut keys, a shortcut key prompt area is added. If the output results of the behavior prediction model indicate that the user prefers mouse operation, the mouse click area layout is optimized.

[0051] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent interface optimization method based on user behavior analysis, characterized in that: The method comprises: Acquire user operation data, which includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information; Performing cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations; For each of the user groups, extract an operation feature vector, wherein the operation feature vector includes a shortcut key combination frequency, a mouse click position distribution, and a drag operation duration distribution; Constructing a user feature matrix according to the operation feature vector; A random forest algorithm is used to train a behavior prediction model according to the user feature matrix, wherein the behavior prediction model is used to predict the user's next operation tendency, and the user's next operation tendency includes the probability of using shortcut keys and the mouse click area preference; Obtaining an output result of the behavior prediction model; The interface layout is dynamically adjusted according to the output result of the behavior prediction model. If the output result of the behavior prediction model shows that the user prefers to use shortcut keys, a shortcut key prompt area is added. If the output result of the behavior prediction model shows that the user prefers mouse operation, the mouse click area layout is optimized.

2. The method according to claim 1, characterized in that The obtaining of user operation data, wherein the user operation data includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information, includes: Obtain user operation behavior data when using the software, including keyboard shortcut combinations, mouse clicks and drags, and other operations, as well as the timestamps and coordinate location information corresponding to these operations; For the acquired user operation behavior data, data preprocessing technology is used to clean and standardize the data, remove noise data, normalize the timestamp and coordinate position information, and obtain standardized user operation behavior data; According to the preset operation behavior pattern library, the standardized user operation behavior data is subjected to feature extraction to obtain the feature vector of the user operation behavior, including the operation type, operation time, operation location and other features; A clustering algorithm is used to perform cluster analysis on the feature vectors of user operation behaviors, and user operation behaviors are divided into different operation modes according to the clustering results, such as shortcut key operation mode, mouse click operation mode, and drag operation mode. For different operation modes, corresponding user behavior prediction models are constructed respectively, such as the shortcut key operation prediction model based on the sequence model and the mouse click operation prediction model based on the graph model, etc. The user behavior prediction model under each operation mode is obtained through training; During the software operation, the user's operation behavior data is obtained in real time, and preprocessed and feature extracted to obtain the feature vector of the current user operation. The corresponding user behavior prediction model is selected according to the operation type to predict the user's subsequent operation behavior; Based on the user behavior prediction results, the software's user interface and interaction methods are dynamically adjusted. For example, commonly used shortcut keys are automatically recommended based on the user's shortcut key operation habits, and button layout is automatically optimized based on the user's mouse click position, etc., to improve user operation efficiency and experience.

3. The method according to claim 1, characterized in that The cluster analysis of the user operation data is performed to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of the drag operation, including: Obtain user operation data during the use of the software, including indicators such as the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations; Preprocess the acquired user operation data, remove outliers and missing values, and normalize the data to make different indicators comparable; Based on the preprocessed user operation data, the K-means clustering algorithm is used to perform cluster analysis on users, and users with similar operation behaviors are divided into the same group; By using methods such as the elbow rule or silhouette coefficient, the optimal number of clusters is determined to obtain different user groups; For each user group, analyze their operation behavior characteristics and summarize the common characteristics and personalized needs of the group; Design targeted personalized service solutions based on the characteristics and needs of different user groups, such as providing a list of commonly used shortcut keys and optimizing mouse click operation processes; Implement personalized service solutions in the software. When users use the software, corresponding personalized services are provided based on the group they belong to, thereby improving user experience and efficiency.

4. The method according to claim 1, characterized in that: For each user group, extract an operation feature vector, wherein the operation feature vector includes shortcut key combination frequency, mouse click position distribution, and drag operation duration distribution, including: Acquire data of multiple pre-divided user groups, and extract operation feature vectors for each user group; Using statistical analysis methods, the frequency distribution of shortcut key combinations used by each user group is calculated to obtain the shortcut key combination frequency feature vector; Through image processing algorithms, cluster analysis is performed on the mouse click position coordinate data of each user group to obtain the mouse click position distribution feature vector; According to the start and end timestamps of the user's drag operation, the drag operation duration data distribution of each user group is calculated to obtain the drag operation duration distribution feature vector; Combine feature vectors such as shortcut key combination frequency, mouse click position distribution, and drag operation duration distribution to form a user group operation feature vector; Use support vector machine algorithm to classify user group operation feature vectors and train user group operation behavior classification model; If the matching degree between the operation feature vector of the new user and the operation feature vector of a certain user group exceeds a preset threshold, it is determined that the new user belongs to the corresponding user group.

5. The method according to claim 1, characterized in that The step of constructing a user feature matrix according to the operation feature vector comprises: Obtain the user's operation log data on the application or platform, extract operation features based on indicators such as operation type, operation frequency, and operation duration, and generate an operation feature vector; For each user, the corresponding operation feature vectors are combined to construct the feature matrix of the user. Each row of the matrix represents an operation feature vector, and each column represents a feature dimension. Perform dimensionality reduction processing on the user feature matrix, use the principal component analysis (PCA) algorithm to extract the main features of the matrix, reduce the feature dimension, and obtain the compressed user feature matrix; According to the compressed user feature matrix, clustering algorithms (such as K-means) are used to group users, and users with similar operation behavior patterns are divided into the same group to characterize the behavioral characteristics of different user groups; In each user group, analyze the user's preference characteristics based on the user's operation type and operation object, build a user preference portrait, and characterize the user's interest preferences in different business scenarios; Based on the behavior patterns and preference characteristics of different user groups, association rule mining algorithms are used to discover the association rules and patterns between the operation behaviors and preference characteristics of different user groups for business decision-making and personalized recommendations; Regularly update user operation log data, recalculate user feature matrix and user group segmentation, dynamically update user portraits, and continuously optimize business strategies and user experience.

6. The method according to claim 1, characterized in that The random forest algorithm is used to train a behavior prediction model according to the user feature matrix, and the behavior prediction model is used to predict the user's next operation tendency, and the user's next operation tendency includes the probability of using shortcut keys and the mouse click area preference, including: Obtain the user's historical operation records, extract user features, and build a user feature matrix. The features include the frequency of shortcut key usage and the distribution of mouse click locations, etc. The random forest algorithm is used to train the behavior prediction model with the user feature matrix as input. The model output is the probability distribution of the user's next operation tendency. According to the user's current operation status, the user characteristics are obtained and input into the trained behavior prediction model to obtain the probability distribution of the user's next operation tendency; Obtain the probability values ​​of shortcut key usage probability and mouse click area preference from the operation tendency probability distribution, and compare them with the preset shortcut key usage threshold and click area threshold respectively; If the shortcut key usage probability is greater than the threshold, it is determined that the user tends to use shortcut keys, and the shortcut key that the user is most likely to use is predicted based on the historical shortcut key usage records; If the probability of mouse click area preference is greater than the threshold, it is judged that the user tends to click the mouse, and the area that the user is most likely to click is predicted based on the historical click location distribution; Return the predicted shortcut keys and click area preferences for interface optimization and interaction guidance to improve user experience.

7. The method according to claim 1, characterized in that The obtaining the output result of the behavior prediction model comprises: Obtaining an output result of the behavior prediction model, the output result including the user's behavior intention and user preference; Determine the current needs of the user according to the user behavior intention in the output result, and select service content that matches the user needs from a preset service content library according to the current needs of the user; According to the user preferences in the output results, further screening out personalized service contents matching the user preferences from the service contents screened out in the service content library; Determine recommended content to recommend to the user based on the personalized service content, and push the recommended content to the user terminal for display; Obtaining user feedback information on the recommended content, and adjusting the parameters of the behavior prediction model according to the user feedback information, so that the behavior prediction model can more accurately predict the user's behavior intention and preference; When adjusting the parameters of the behavior prediction model, a gradient descent algorithm is used to optimize the parameters of the behavior prediction model, and the prediction accuracy of the behavior prediction model is continuously improved through multiple iterations; The optimized behavior prediction model is applied to subsequent user behavior prediction and personalized recommendation processes to continuously improve the accuracy of recommended content and user experience.

8. The method according to claim 1, characterized in that: The method dynamically adjusts the interface layout according to the output result of the behavior prediction model, and if the output result of the behavior prediction model indicates that the user prefers to use shortcut keys, increases the shortcut key prompt area, and if the output result of the behavior prediction model indicates that the user prefers mouse operation, optimizes the mouse click area layout, including: Obtain the user's historical operation behavior data, including the frequency and mode of using shortcut keys and mouse operations, as training data for the behavior prediction model; Use machine learning algorithms such as decision trees and random forests to build a user behavior prediction model, train it based on the user's historical operation behavior data, and obtain a trained behavior prediction model; Collect user operation behavior data in real time while the user is using the software, input it into the behavior prediction model, obtain the output of the model, and judge the user's operation preference; If the model output results show that the user prefers to use shortcut keys, the area of ​​the shortcut key prompt area in the interface is dynamically increased, and the layout design of the shortcut keys is optimized, such as arranging commonly used shortcut keys in a prominent position; If the model output results show that the user prefers mouse operation, the layout design of the mouse click area in the interface is dynamically optimized, such as increasing the button size, optimizing the button position arrangement, etc., to improve the convenience of mouse operation; Continuously track user operation behaviors, regularly update behavior prediction models, dynamically adjust interface layout based on the latest prediction results output by the model, and achieve adaptive optimization of the interface; Evaluate the user experience of the optimized interface through A / B testing and other methods, collect user feedback, continuously improve behavior prediction models and interface layout optimization strategies, and enhance the overall user experience.

9. An intelligent interface optimization system based on user behavior analysis, characterized in that: The system comprises: A data acquisition module is used to acquire user operation data, wherein the user operation data includes shortcut key combinations, mouse clicks, drag operations and their corresponding timestamps and coordinate position information; A cluster analysis module, used to perform cluster analysis on the user operation data to obtain different user groups, wherein the cluster analysis is performed based on the frequency of shortcut key usage, the number of mouse clicks, and the duration of drag operations; A feature extraction module, for extracting an operation feature vector for each user group, wherein the operation feature vector includes a shortcut key combination frequency, a mouse click position distribution, and a drag operation duration distribution; A matrix construction module, used to construct a user feature matrix according to the operation feature vector; A model training module, used to train a behavior prediction model based on the user feature matrix using a random forest algorithm, wherein the behavior prediction model is used to predict the user's next operation tendency, and the user's next operation tendency includes a shortcut key usage probability and a mouse click area preference; A result acquisition module, used to obtain the output result of the behavior prediction model; The interface adjustment module is used to dynamically adjust the interface layout according to the output results of the behavior prediction model. If the output results of the behavior prediction model indicate that the user prefers to use shortcut keys, a shortcut key prompt area is added. If the output results of the behavior prediction model indicate that the user prefers mouse operation, the mouse click area layout is optimized.

Images