Research And Implementation Of The Online Registration Recommended System Based On Web Usage Mining

Along with the continuous development of Internet technology, e-commerce has also made great development. People enjoy the convenience of electronic commerce also have to face the e-commerce site on the commodities continue to increase, they need to find products more difficultly. In addition, if businesses want to gain advantage in e-commerce, they must do a better grasp of the characteristics of customer and market trends. This information can be based on historical customers on the web server log file analysis to be left behind. But every mouse click from visitors will be left a record in the log file. A larger e-commerce site can generate millions of records for a day. Faced with such a huge amount of data, using artificial means impossibly to derive any information!In order to take advantage of these data, we use the Web data mining technology to process these data. Using user's patterns of the excavations, we can get the user's access patterns. Using clustering technology, we can cluster visitors. And then according to user clicks on habits in a cluster, choosing the best-selling product in this cluster recommend to other users. Using the Apriori algorithm to find association rules in the database, and then based on the user's current shopping behavior to predict, take the initiative to recommend to improve the user purchases on the site several times odds.Starting from the data mining, this paper began to discuss the data mining technology research background, meaning and the development of the status quo at home and abroad, and then for the specific circumstances of e-commerce, specifically studied mining technology in electronic commerce applications. In the research process, we mainly research the significance of the association rules in data mining, as well as the significance and implementation of the classic Apriori algorithm for association rules algorithm; the common techniques and analysis methods in web data mining; as well as the architecture of browser-based e-commerce recommendation system and the implement of the internal algorithms. Obtained the knowledge model from Web mining, we can provide customers with personalized referral service through e-commerce recommendation system that can improve the site's sense of belonging and customer satisfaction at the same time can increase the competitiveness of the site; Business e-commerce site based on the information received, can better grasp of market dynamics. They can better grasp the market dynamics, and make better decisions to demand market demand.