An Effective Fuzzy Clustering Algorithm in Web Mining

摘要

This paper presents an effective method for clustering web transactions from web logs. Each web transaction is denoted as a fuzzy vector with the same length. Each element of the fuzzy vector is a fuzzy linguistic variable representing fuzzy time duration on a web page. Then we defined the relationship between any two web transactions and added them into fuzzy vector. Furthermore, a rough c-means algorithm is adopted to cluster these new formed fuzzy vectors. Analysis and comparison of the method demonstrate the given clustering algorithm is more meaningful and effective.