Automatic clustering method of multivariate data using Gaussian Dirichlet process mixture model

Abstract

Finite mixture models have largely been used for providing a convenient format framework for clustering and classification for multivariate data. But most of these models assume that the number of components in mixture model is known in advance. To resolve this issue, we introduce a novel nonparametric Bayesian clustering model, is called Gaussian Dirichlet process mixture model, for the automatic clustering algorithm of multivariate data, and we have also described an efficient variational Bayesian inference algorithm for the proposed model. We apply it to a series of various clustering problems, demonstrating its advantages over existing methodologies.