Research On Personalized Recommendation System Based On Intelligent Perception Analysis User Behavior

In recent years,the rapid development of computer science and technology,promote the penetration of Internet technology in people's lives.People in their daily lives are constantly via the Internet to obtain information access,query and a variety of network services.However,during the rapid development of Internet technology,network security has become an important factor hindering its development,and security issues have also brought great challenges to the development of e-commerce technology.With the development of e-commerce technology,more and more enterprises begin to in-depth analysis the behavior of network users,to accurately grasp the user's hobbies by analyzing the user's behavior,so according to that,enterprises can complete the personalized product recommended services.In this process,the user can fully improve the transaction success rate,while enterprises can improve the profitability.Many government and non-profit sites are also aware of the importance of analyzing user behavior for network maintenance and development,so more and more scientific research workers begin to invest in the study of network user behavior.It is becoming more and more important to analyze the behavior of network users.Many e-commerce sites are through the user's click behavior to carry out the user behavior data collection.With the continuous development of Internet technology,more and more advertising is full of the network.All users in the process of using the Internet,their own browsing records will be automatically identified and recorded by the network,which also gives the user behavior analysis provides a very accurate basis for the behavior.In the study of user behavior,the data collection and site data extraction is very important.The research of this topic is based on the intelligent perception technology to analyze the user behavior accurately.In this paper,the concept of network user behavior and related technical core are analyzed,at the same time according to the user behavior of the different classification,it can effectively improve the accuracy of user behavior analysis.The user's behavior is modeled to understand the input andoutput of the model,thus improving the understanding of the user behavior analysis system model.At the same time,based on the content analysis and the collaborative filtering algorithm,the measurement items are added to the user behavior analysis,and the mean square error is used to improve the accuracy of the calculation.Through the system test can be found,the use of weighted combination algorithm can effectively reduce the time complexity of the algorithm,which effectively compensates the cold start problem in the collaborative filtering algorithm and greatly improve the accuracy of the personalized recommendation system.