Fuzzy Clustering in Web Mining

Abstract

Web mining is the use of data mining techniques to automatically discover and extract information from web. Clustering is one of the possible techniques to improve the efficiency in information finding process. Conventional clustering classifies the given data objects into exclusive clusters. However such a partition is insufficient to represent many real situations. Hence a fuzzy clustering method is offered to construct clusters with uncertain boundaries and allows the object to belong to multiple clusters with degree of membership. Web data has fuzzy characteristics, so fuzzy clustering is better suitable for web mining in comparison with conventional clustering. In this paper, we have proposed two algorithms that are Fuzzy c-Means (FCM) and Clustering based on Fuzzy Equivalence Relations which can be used for web page mining and web usage mining. The results obtained from the proposed algorithm are more convincing. The experimental results are carried out on different algorithmic parameters on real data. The analysis is being done by comparing the proposed algorithm with conventional clustering algorithms.