期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2011
卷号：2
期号：3
页码：965-967
出版社：TechScience Publications
摘要：Web mining a hot research area of data mining can be classified into three main areas: Web Usage Mining, Web content Mining and Web Structure Mining. Web usage mining is a kind of web mining, which exploits data mining techniques to discover valuable information from navigation behavior of World Wide Web users. There are generally three tasks in Web Usage Mining: Preprocessing, Pattern analysis and Knowledge discovery. Preprocessing cleans log file of server by removing log entries such as error or failure and repeated request for the same URL from the same host etc... The main task of Pattern analysis is to filter uninteresting information and to visualize and interpret the interesting pattern to users. The statistics collected from the log file can help to discover the knowledge. This knowledge collected can be used to take decision on various factors like Excellent, Medium, Weak users and Excellent, Medium and Weak web pages based on hit counts of the web page in the web site. The design of the website is restructured based on user’s behavior or hit counts which provides quick response to the web users, saves memory space of servers and thus reducing HTTP requests and bandwidth utilization. This paper addresses challenges in three phases of Web Usage mining along with Web Structure Mining.duc
关键词：web mining; web site; hit count; log file; HTTP;Topology; URL.
