Study On E-Commerce Recommendation System Based On Web Mining

With the popularity of the Internet and e-commerce development, e-commerce system provides users with more choices, at the same time, its structure has become more complex, users are often lost in a large number of goods in the information space, and could not find their own needs. In the increasingly fierce competitive environment, the recommendation system can effectively retain customers and prevent the loss of customers, as well as increase sales of e-commerce enterprises and its competitiveness.Recommendation system in e-commerce has good prospects for the development and application, which has gradually become an important research in e-commerce technology, but with the further expand to the scale of goods recommendation system ,it is also facing a series of challenges. Aiming at the major challenges which Recommendation system faces, this paper does a useful exploration and research in the following three areas in e-commerce recommendation system.Firstly, the article analyzes the characteristics of data mining technology and the web mining's superiority and its important role in e-commerce. Then introduces and analyze the recommendation systems in the e-commerce and its work process.Second, this paper gives the architecture framework of e-commerce recommendation system. After the concept of e-commerce recommendation system, this paper does some research on the processes and critical technologies from the various stages of e-commerce recommendation system including the log preprocessing, model patterns discovering, researching and its application.Thirdly, this paper gives a suitable algorithm for Web log mining. Aiming at the real-time challenge which recommendation system faces, this paper uses the fuzzy clustering technology to improve Markov model which can be used to analyze web data. Following the description and characteristics of the algorithm, the advantages and disadvantages of the algorithm and improve the direction are given.However, the model still has many deficiencies in the room for improvement: in particular, accuracy and efficiency issues. Improving the efficiency of Algorithm is an important issue in the area of current research .As the developing in Web log data analysis and research; we believe that e-commerce personalized recommendation service will be practical.