Abstract
A Gaussian mixture model is a weighted sum of parametric Gaussian components. These parametric density functions are widely used in data mining and pattern recognition. In this work we propose an efficient method to model a density function as a Gaussian mixture through an iterative algorithm that allow us to estimate the parameters of the model for a given data set. For this purpose we use the Gini Index, a measure of the inequality degree between two probability distributions. The Gini Index is obtained by finding the solution of an optimization problem. Our model consists in minimizing the Gini Index between an empirical distribution and a parametric distribution that is a Gaussian mixture. We will show some simulated examples and real data examples, with two widely used datasets, to observe the efficiency and properties of our model.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bassetti, F., Bodini, A., Regazzini, E.: On minimum Kantorovich distance estimators. Stat. Probab. Lett. 76(12), 1298–1302 (2006)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, New York (2006)
Capinski, M., Kopp, P.: Measure, Integral and Probability. Springer Undergraduate Mathematics Series. Springer, London (2013). https://books.google.com.mx/books?id=5d6PBAAAQBAJ
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. Roy. Stat. Soc.: Ser. B (Methodol.) 39(1), 1–22 (1977)
Giorgi, G.M., Gigliarano, C.: The Gini concentration index: a review of the inference literature. J. Econ. Surv. 31(4), 1130–1148 (2017)
Greenspan, H., Ruf, A., Goldberger, J.: Constrained Gaussian mixture model framework for automatic segmentation of MR brain images. IEEE Trans. Med. Imaging 25(9), 1233–1245 (2006)
Meng, X.L., Rubin, D.B.: On the global and component wise rates of convergence of the EM algorithm. Linear Algebra Appl. 199, 413–425 (1994)
Povey, D., et al.: The subspace Gaussian mixture model–a structured model for speech recognition. Comput. Speech Lang. 25(2), 404–439 (2011)
Rachev, S.T., Klebanov, L.B., Stoyanov, S.V., Fabozzi, F.: The Methods of Distances in the Theory of Probability and Statistics. Springer, New York (2013). https://doi.org/10.1007/978-1-4614-4869-3
Reynolds, D.A.: Gaussian mixture models. In: Encyclopedia of Biometrics, vol. 741 (2009)
Rubner, Y., Tomasi, C., Guibas, L.J.: The earth mover’s distance as a metric for image retrieval. Int. J. Comput. Vis. 40(2), 99–121 (2000). https://doi.org/10.1023/A:1026543900054
Torres-Carrasquillo, P.A., Reynolds, D.A., Deller, J.R.: Language identification using gaussian mixture model tokenization. In: 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 1, p. I-757. IEEE (2002)
Ultsch, A., Lötsch, J.: A data science based standardized Gini index as a Lorenz dominance preserving measure of the inequality of distributions. PloS One 12(8), e0181572 (2017)
Vaida, F.: Parameter convergence for EM and MM algorithms. Statistica Sinica 15, 831–840 (2005)
Villani, C.: Topics in Optimal Transportation. American Mathematical Society, Providence (2003)
Xu, L., Jordan, M.I.: On convergence properties of the EM algorithm for Gaussian mixtures. Neural Comput. 8(1), 129–151 (1996)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
López-Lobato, A.L., Avendaño-Garrido, M.L. (2020). Using the Gini Index for a Gaussian Mixture Model. In: Martínez-Villaseñor, L., Herrera-Alcántara, O., Ponce, H., Castro-Espinoza, F.A. (eds) Advances in Computational Intelligence. MICAI 2020. Lecture Notes in Computer Science(), vol 12469. Springer, Cham. https://doi.org/10.1007/978-3-030-60887-3_35
Download citation
DOI: https://doi.org/10.1007/978-3-030-60887-3_35
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-60886-6
Online ISBN: 978-3-030-60887-3
eBook Packages: Computer ScienceComputer Science (R0)