A fuzzy and locally sensitive method for cluster analysis
A.C.G. Thome · 2002
Cluster analysis has been playing an important role in pattern recognition, image processing and time series analysis. The majority of the existing clustering algorithms depend on initial parameters and assumptions about the underlying data structure. A fuzzy method of mode separation is proposed. The method addresses the task of multi-modal partitioning through a sequence of locally sensitive searches guided by a stochastic gradient ascent procedure, and addresses the cluster validity problem through a global partition performance criterion. The algorithm is computationally efficient and provides good results when tested with a number of simulated and real data sets.