A Survey of Novel Clustering and Knowledge Extraction from Log

Now, a days the explosive growth of Internet is rapidly increasing and knowledge extraction become the objective of most business like advertisement. Due to the biggest source of information the information available to the user is not relevant. to overcome this problem we have to understand the user need and preferences. Web mining is nothing but mining the information related to web users. this is crucial task to get web users interest and need. In web mining there are several task that has to be perform such as preprocessing, clustering and knowledge extraction. we use clustering technique in this paper for grouping web users having same interest. There are different clustering systems that need to be survey. The knowledge extraction deals with extracting users interest through their preferences. here in this paper we are using log files as a source for extracting information. In the existing system k-means clustering, priori algorithm are used. A preprocessing identifies each user through his transaction logged in file. World Wide Web is intermediate of web pages and links. In web usage mining is also the application of data mining techniques to determine usage pattern from web data for understanding the needs.