An Effective method for Web Log Preprocessing and Page Access Frequency using Web Usage Mining

摘要

World Wide Web is rising rapidly and enormous amount of information is produced due to user's communications with web sites. To exploit this information, recognizing usage pattern of users is very significant. Web Usage Mining is the application of data mining techniques to find out the useful, hidden information about the users and interesting patterns from data extracted from Web Log files. It supports to know frequently accessed pages, suppose user navigation, progress web site structure etc. In command to relate Web Usage Mining, variety of actions is executed. This paper describes the procedure of Web Usage Mining consisting steps: Data Collection, Pre-processing, Pattern Discovery and Pattern Analysis. It has furthermore existing several approaches for example statistical analysis; clustering, association rules and sequential pattern are individual used to determine patterns in web usage mining In this paper describes data preprocessing technique of web usage mining, after completion of data preprocessing, any kind of irrelevant information can be sort out. We have also proposed an algorithm and its implementation for web log preprocessing in web usage mining. Every page has been allocated with an individual token. According to this token and frequency, data mining technique (Classification, Association Rules, and Clustering) can be applied. In this article we can simply discover the highest and lowest value according to page access frequency.