Abstract
This paper compares the performance of Generalised Fisher Ratio, Normalised Entropy and Functional Merging on assessing the exact parameters and number of components of finite Gaussian mixtures. The three methods use different statistical criteria for calculating the right number of components and have in common the Expectation-Maximisation algorithm for assessing the parameters of Gaussian mixtures. After applying the three above criteria to a similar benchmark as in paper (Celeux and Soromenho, Journal of Classification, 13, 1996) we show that Functional Merging surpass the performance of the other two methods.
Similar content being viewed by others
References
R. Reed, “Pruning Algorithms–A Survey,” IEEE Transactions on Neural Networks, vol. 4, no.5, Sept. 1993, pp. 740–747.
S.E. Fahlman and C. Lebiere, “The Cascade-Correlation Learning Architecture,” in Neural Information Processing Systems 2, D.S. Touretzky (Ed.), Morgan-Kaufmann, 1990, pp. 524–532.
H. Akaike, “A New Look at the Statistical Identification Model,” IEEE Transactions onAutomatic Control, vol. 19, 1974, pp. 716–723.
G. Schwarz, “Estimating the Dimension of a Model,” Annals of Statistics, vol. 6, 1978, pp. 461–464.
H. Bozdogan, “Choosing the Number of Component Clusters in the Mixture Model Using a New Information Complexity Criterion of the Inverse-Fisher Information Matrix,” in Information and Classification, O. Opitz and B. Lausen (Eds.), Heidelberg: Springer-Verlag, 1993, pp. 40–54.
S. Richardson and P.J. Green, “On Bayesian Analysis of Mixtures with an Unknown Number of Components,” Journal of Royal Statistical Society B, vol. 59, no.4, 1997, pp. 731–792.
C.G. Molina, “Model Adaptation of Normal Mixtures by Functional Merging on Probability Density Estimation,” Tech. Rep. CUED/F-INFENG/ 1996, Cambridge University Engineering Department, Trumpington Street, Cambridge CB2 1PZ, England, Sept. 1996.
S. Zheng and C.G. Molina, “Measuring Tree-Ring Parameters Using the Generalised Fisher Ratio,” in EUSIPCO, Rhodes, Greece, Sept. 8-11, 1998.
G. Celeux and G. Soromenho, “An Entropy Criterion for Assessing the Number of Clusters in a Mixture Model,” Journal of Classification, vol. 13, 1996, pp. 195–212.
A.P. Dempster, N.M. Laird, and D.B. Rubin, “Maximum Likelihood from Incomplete Data Via the EM Algorithm,”Journal of Royal Statistical Society, vol. B, no.39, 1977, pp. 1–38.
C.M. Stow, A.C.T. Kennington, C.G. Molina, and William J. Fitzgerald, “Experimental Issues of Functional Merging on Probability Density Estimation,” in Artificial Neural Networks, London, Institution of Electrical Engineers, July 7-9, 1997, pp. 123–128.
M. Coates, C.G. Molina, and W. Fitzgerald, “Regionally Optimised Kernels for Time Frequency Distributions,” in ICASSP, International Conference in Acoustics, Speech and Signal Processing (Seatle, Washington, USA), May 12-15, 1998.
R. Duda and P. Hart, Pattern Classification and Scene Analysis, New York: Wilsey, 1973.
S. Krishnan, K. Samudravijaya, and P.V.S. Rao, “Feature-Selection for Pattern-Classification with Gaussian Mixture-Models–A New Objective Criterion,” Pattern Recognition Letters, vol. 17, no.8, 1996, pp. 803–809.
S. Zheng and C.G. Molina, “Automatic Count of Neural Cells Using the Generalised Fisher Ratio,” in NNESMED'98, Pisa, Italy, Sept. 1998.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Molina, C.G. Assessing the Number of Components in Finite Gaussian Mixtures by Generalised Fisher Ratio, Normalised Entropy Criterion and Functional Merging. The Journal of VLSI Signal Processing-Systems for Signal, Image, and Video Technology 26, 95–103 (2000). https://doi.org/10.1023/A:1008199518065
Published:
Issue Date:
DOI: https://doi.org/10.1023/A:1008199518065