EGMM: An evidential version of the Gaussian mixture model for clustering

The Gaussian mixture model (GMM) provides a simple yet principled framework for clustering, with properties suitable statistical inference. In this paper, we propose new model-based clustering algorithm, called EGMM (evidential GMM), in the theoretical of belief functions to better characterize cluster-membership uncertainty. With mass function representing cluster membership each object, evidential distribution composed components over powerset desired clusters is proposed entire dataset. parameters are estimated by specially designed Expectation–Maximization (EM) algorithm. A validity index allowing automatic determination proper number also provided. as classical GMM, but can generate more informative partition considered synthetic and real dataset experiments show that performs than other representative algorithms. Besides, its superiority demonstrated an application multi-modal brain image segmentation.