A Weighted Fuzzy Clustering Algorithm for Data Stream

Abstract

Mining data streams poses great challenges due to the limited memory availability and real time query response requirement. One of the most important mining tasks is clustering. There already lots of clustering algorithms for data stream have been presented. Fuzzy cluster is an important clustering method. However, to the best of our knowledge, all the clustering algorithms are hard clustering methods, fuzzy clustering algorithm is presently not used directly for data streams. Fuzzy c-means (FCM) is a typical fuzzy clustering algorithm. In this paper, we extend FCM and propose a weighted fuzzy algorithm for clustering data stream. Experimental results on both synthetic and real data sets show its superiority over the traditional FCM algorithms.