论文信息 - On Fitting Finite Dirichlet Mixture Using ECM and MML

On Fitting Finite Dirichlet Mixture Using ECM and MML

Gaussian mixture models are being increasingly used in pattern recognition applications. However, for a set of data other distributions can give better results. In this paper, we consider Dirichlet mixtures which offer many advantages [1]. The use of the ECM algorithm and the minimum message length (MML) approach to fit this mixture model is described. Experimental results involve the summarization of texture image databases.

Nizar Bouguila | Djemel Ziou | D. Ziou | N. Bouguila

[1] Geoffrey J. McLachlan,et al. Finite Mixture Models , 2019, Annual Review of Statistics and Its Application.

[2] James O. Berger,et al. Ockham's Razor and Bayesian Analysis , 1992 .

[3] C. S. Wallace,et al. MML mixture modelling of multi-state, Poisson, von Mises circular and Gaussian distributions , 1997 .

[4] Jonathan J. Oliver,et al. Finding overlapping components with MML , 2000, Stat. Comput..

[5] Michael Unser,et al. Sum and Difference Histograms for Texture Classification , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6] Robert M. Haralick,et al. Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[7] David L. Dowe,et al. MML clustering of multi-state, Poisson, von Mises circular and Gaussian distributions , 2000, Stat. Comput..

[8] Nizar Bouguila,et al. Unsupervised learning of a finite mixture model based on the Dirichlet distribution and its application , 2004, IEEE Transactions on Image Processing.

[9] C. E. SHANNON,et al. A mathematical theory of communication , 1948, MOCO.

[10] Xiao-Li Meng,et al. Maximum likelihood estimation via the ECM algorithm: A general framework , 1993 .