A clustering algorithm based on minimum volume
Raghu J. Krishnapuram, Jong Woo Kim · Proceedings of IEEE 5th International Fuzzy Systems · 2002
Most fuzzy clustering algorithms are derived from the fuzzy C-means (FCM) algorithm, which minimizes the sum of squared distances from the prototypes weighted by the corresponding memberships. In this paper, we consider a new clustering algorithm based on the minimization of the sum of the volumes of the clusters. The performance of the algorithm is shown to be better than that of the traditional algorithms when the data set contains clusters of widely varying sizes, shapes, and densities.