Research & Implement On Mining Logs Of Website

Web site logs analysis and research system based on data mining uncovers the hidden regulations among the interactive data between a web server and its users in order to improve service of web and enhance security. According to data mining, this system aims to find normal and abnormal access patterns and interested paths, then supply better and personalized services and give advices to administrator to enhance security of web site. The data resources of this system are web logs and web site topology. We pay more attention upon similarity analysis algorithms on mining browsing pattern and make a comparation between classical similarity algorithms with our proposal one. The comparation tells us the optimized similarity algorithm is advantage. Then introduce a new way to find user's interested path upon the optimized algorithm large graph from classical Apriori algorithms. Total system is a base of the next step to realize web site intelligentization.This thesis is composed of the following seven chapters.Chapter 1: This chapter chiefly introduces the data mining technology and current status in China and foreign countries. It comes out the main content and significations. Chapter 2: This chapter describes data mining in detail and the facing difficulties because of different data constructions. It also introduces log data and procedure to mining logs. Chapter 3: How to collecting data for logs mining and its channels and flow charts are discussed in this chapter. Chapter 4: In this chapter preprocessing phase is introduced from the target, procedure and algorithm of data cleaning and how to build database. It also focuses on user identification, path identification and affair identification and so on. Chapter 5: This chapter discusses several logs mining algorithms including classifying logs data into two parts: normal and abnormal data to enhance net security by calculating session similarity, finding interested path by large graph algorithm like Apriori.Chapter 6: In this chapter there conclude into a system structure of logs mining and propose partial experiment results. Chapter 7: In this chapter we review our research and propose conclusions including advantage and disadvantage points.