Abstract:

Associating feature with weights for each cluster is a common approach in clustering algorithms and determining the weight values is crucial in generating valid partition. This paper introduces a novel method in the framework of granular computing that incorporates fuzzy sets, rough sets, and shadowed sets, and calculates feature weights at each iteration automatically. The method of feature weighting can realize the clustering objective more effectively, and the clustering validity indices of DB, Dunn and XB are applied to analyze the validity of partition-based clustering. Comparative experiments results reported for real data sets illustrate that the proposed algorithms are always convergent and more effective in handing overlapping among clusters and more robust in the presence of noisy data and outlier.