Research And Implementation Of Web User Experience Optimization Based On Web Page Reconstruction

With the continuous development of mobile Internet technologies,more and more content is available on web pages,and the number of web page fonts is increasing.This poses a huge challenge for browsers to quickly load web pages.At present,there are many web browsing experiences(Web QoE)for measuring web page loads.The most popular research is to increase the page's Above the Fold Time(AFT).This article uses AFT as a research index for measuring web browsing experience in existing webpage loading,which is different from the overall loading time,PLT).Due to the popularity and development of mobile devices,it is limited by the screen size of mobile devices,and AFT is more able to compare to the overall loading duration.Visually respond to user experience.In the current research,there are technologies that can reduce AFT.However,there are still some deficiencies in this type of technology:1)The current work reduces the AFT by increasing the computing power of the hardware and the bandwidth of the network.It is impossible to model the factors affecting the AFT.Analysis.2)There is no mature AFT measurement system that does not affect users' normal browsing.3)The current AFT optimization system is too complicated and difficult to use in the terminal.We aim at the defects in the above research,a new lightweight AFT optimization system based on web page refactoring is proposed.This system can bring obvious optimization to the AFT of webpage loading.This paper mainly has three aspects:1)Web QoE modeling theoretically analyzes the key factors affecting web page AFT.2)For AFT measurement,design and implement a lightweight tool for web page AFT measurement without affecting the user's browsing experience.3)Design and implement a set of lightweight The first screen loading time optimization system can bring AFT optimization to webpage loading.In the system design and implementation,this article mainly has three innovations:1)Analyze the key factors affecting the loading time of the first screen from a theoretical perspective,design and implement a lightweight first-time load-time measurement tool;2)Focus on the critical path The implementation of JavaScript and CSS files,the introduction of machine learning classification algorithm predicts the impact of JavaScript execution on the first screen,based on the prediction results to adjust the structure of the web page;for the existence of the style of the CSS file is deprecated,the optimization of the CSS file;The first screen loading optimization system design for web page refactoring 0.3)The first screen load time optimization system based on web page reconstruction is implemented.Finally,this paper tests the system from two aspects:the prediction accuracy of the system's prediction model and the optimization of the system under different network conditions.Among them,for the test of the prediction model,this paper adopts a 10-fold cross-validation method to test the model from different feature dimensions.The results show that the prediction accuracy of this model is as high as 95%,and it can well complete the task of prediction.For the system optimization space test,the results show that when the network conditions are excellent,the AFT after system optimization is reduced by an average of 30.3%;when the network conditions are in a good and general state,the AFT after the system optimization is reduced by an average of 25%;in the poor network conditions At the time,the AFT after system optimization decreased by an average of 19.6%;PLT decreased by an average of 10%.