Sequential Patterns Mining And Application In Web Log

With the rapid development and application of Internet, the resource of Web log is becoming more and more abundant. The prominent problem is how to analyze and use the great amount of data. The Web log mining is a new technique of Web information processing, and it is also an important application of data mining in the Internet domain. With the rapid development of Internet, web log mining is widely applied to E-commerce and individuating Web. We can improve the structures of Web sites and the system design of Web application, monitor the server and provide individual service to the users. In addition, Web log mining could optimize the design of the Web sites and improves the decision-making of the market by analyzing access path of the users who use the Web mining.Web log mining is to use the idea of data mining to analyze and deal with the server logs, thus we can use these techniques to realize the above functions. However, the algorithms of mining user's visiting paths used presently are mostly to use the Apriori algorithm which mines directly frequent itemset of Boolean association rule. It ignores how to combine with the features of the access path to improve the algorithms , which can get better mining results and make mining more efficient. Based on the research of data mining technique, we focus on the feature, method and related techniques of Web log mining. The process of the log processing and several kinds of efficient methods of data pretreatment are discussed, and the identification of the many unique users and user session from sever logs is realized . Association rule and sequential pattern techniques are the methods that study and find the relationship in the items of transaction database. In this paper, the mining techniques of association rule and sequence pattern recognition techniques are researched, moreover, sequential pattern recognition algorithm based on association rules is realized which make use of the advantages of the two techniques. The algorithm can extract the information of the user and data from Web sever log, and can identify the user's access path efficiently. The Web log mining techniques is applied to E-commercedomain, Some directions of application and the research of the algorithm which canimprove designs of Web sites are emphasized.The experiment shows that the techniques of Web log mining could efficiently recognize user's access path, and valuable information for the web site administrators and the merchants' decision-making is provided, which realize the service of individuating Web.