Generalized Adaptive Possibilistic C-Means Clustering Algorithm
Spyridoula D. Xenaki, Konstantinos Koutroumbas, Athanasios A. Rontogiannis · 2018
In this paper a generalized possibilistic c-means clustering algorithm, called Generalized Adaptive Possibilistic C-Means (GAPCM), is presented. The algorithm extents the abilities of its ancestor, Adaptive Possibilistic C-Means (APCM), allowing the study of cases where the data form compact and hyper-ellipsoidally shaped clusters, whose points may lie around certain subspaces in the feature space. In addition, these clusters may be located very close, or even intersect each other. More specifically, a proper definition and an adaptivity concept of the parameters that GAPCM involves, during its execution, renders the algorithm able to unravel on its own the actual hyper-ellipsoidal shape of the clusters formed by the data. The performance of the algorithm is assessed through its comparison with other related algorithms on the basis of both simulated and real data sets.