Web Usage Mining through Efficient Genetic Fuzzy C-Means

In process of knowledge discovery from any web-log dataset, most widely and extensively used clustering algorithm for this purpose is Fuzzy c-means (FCM) algorithm because the data of web-log is unsupervised dataset. Due to sensitivity of FCM, it can be easily trapped in a local optimum, and it is also depends on initialization. In this paper we present use of Genetic algorithm in Fuzzy c-means algorithm to select initial center point for clustering in FCM. The purpose of this paper is to provide optimum initial solution for FCM with the help of genetic algorithm to reduce the error rate in pattern creation.