Performance Comparison of Two Streaming Data Clustering Algorithms

摘要

The weighted fuzzy cmean clustering algorithm (WFCM) and weighted fuzzy cmeanadaptive cluster number (WFCMAC) are extension of traditional fuzzy cmean algorithm to stream data clustering algorithm. Clusters in WFCM are generated by renewing the centers of weighted cluster by iteration. On the other hand, WFCMAC generates clusters by applying WFCM on the data & selecting best K± initialize center. In this paper we have compared these two methods using KDDCUP’99 data set. We have compared these algorithms with respect to number of valid clusters, computational time and mean standard error.