Abstract
Asymmetric Gaussian mixture (AGM) model has been proven to be more flexible than the classic Gaussian mixture model from many aspects. In contrast with previous efforts that have focused on maximum likelihood estimation, this paper introduces a fully Bayesian learning approach using Metropolis-Hastings (MH) within Gibbs sampling method to learn AGM model. We show the merits of the proposed model using synthetic data and a challenging intrusion detection application.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bouguila, N., Ziou, D.: MML-based approach for finite dirichlet mixture estimation and selection. In: Perner, P., Imiya, A. (eds.) MLDM 2005. LNCS (LNAI), vol. 3587, pp. 42–51. Springer, Heidelberg (2005). https://doi.org/10.1007/11510888_5
Bouguila, N., Ziou, D., Hammoud, R.I.: A Bayesian non-Gaussian mixture analysis: application to eye modeling. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2007), 18–23 June 2007, Minneapolis, Minnesota, USA (2007)
Sefidpour, A., Bouguila, N.: Spatial color image segmentation based on finite non-Gaussian mixture models. Expert Syst. Appl. 39(10), 8993–9001 (2012)
Bouguila, N., Amayri, O.: A discrete mixture-based kernel for SVMs: application to spam and image categorization. Inf. Process. Manag. 45(6), 631–642 (2009)
McLachlan, G.J., Peel, D.: Finite Mixture Models. Wiley, Hoboken (2000)
Bouguila, N., Ziou, D., Monga, E.: Practical Bayesian estimation of a finite beta mixture through Gibbs sampling and its applications. Stat. Comput. 16(2), 215–225 (2006)
Elguebaly, T., Bouguila, N.: Bayesian learning of generalized Gaussian mixture models on biomedical images. In: Schwenker, F., El Gayar, N. (eds.) ANNPR 2010. LNCS (LNAI), vol. 5998, pp. 207–218. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-12159-3_19
Elguebaly, T., Bouguila, N.: Background subtraction using finite mixtures of asymmetric Gaussian distributions and shadow detection. Mach. Vis. Appl. 25(5), 1145–1162 (2014)
Bouguila, N., Almakadmeh, K., Boutemedjet, S.: A finite mixture model for simultaneous high-dimensional clustering, localized feature selection and outlier rejection. Expert Syst. Appl. 39(7), 6641–6656 (2012)
Boutemedjet, S., Ziou, D., Bouguila, N.: Model-based subspace clustering of non-Gaussian data. Neurocomputing 73(10–12), 1730–1739 (2010)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B (Methodol.) 39(1), 1–38 (1977)
Bouguila, N., Ziou, D.: On fitting finite Dirichlet mixture using ECM and MML. In: Singh, S., Singh, M., Apte, C., Perner, P. (eds.) ICAPR 2005. LNCS, vol. 3686, pp. 172–182. Springer, Heidelberg (2005). https://doi.org/10.1007/11551188_19
Bouguila, N., Ziou, D., Hammoud, R.I.: On Bayesian analysis of a finite generalized Dirichlet mixture via a metropolis-within-Gibbs sampling. Pattern Anal. Appl. 12(2), 151–166 (2009)
Bouguila, N., Elguebaly, T.: A fully Bayesian model based on reversible jump MCMC and finite beta mixtures for clustering. Expert Syst. Appl. 39(5), 5946–5959 (2012)
Hastings, W.K.: Monte Carlo sampling methods using Markov chains and their applications. Biometrika 57(1), 97–109 (1970)
Bouguila, N.: Bayesian hybrid generative discriminative learning based on finite Liouville mixture models. Pattern Recogn. 44(6), 1183–1200 (2011)
Bourouis, S., Mashrgy, M.A., Bouguila, N.: Bayesian learning of finite generalized inverted Dirichlet mixtures: application to object classification and forgery detection. Expert Syst. Appl. 41(5), 2329–2336 (2014)
Elguebaly, T., Bouguila, N.: Bayesian learning of finite generalized Gaussian mixture models on images. Sig. Process. 91(4), 801–820 (2011)
Bouguila, N., Ziou, D.: A powreful finite mixture model based on the generalized Dirichlet distribution: unsupervised learning and applications. In: 17th International Conference on Pattern Recognition, ICPR 2004, Cambridge, UK, 23–26 August 2004, pp. 280–283. IEEE Computer Society (2004)
Bouguila, N., Ziou, D.: Dirichlet-based probability model applied to human skin detection [image skin detection]. In: 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2004, 17–21 May 2004, Montreal, Quebec, Canada, pp. 521–524. IEEE (2004)
Luengo, D., Martino, L.: Fully adaptive Gaussian mixture metropolis-hastings algorithm. In: IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2013, 26–31 May 2013, Vancouver, BC, Canada, pp. 6148–6152. IEEE (2013)
Tavallaee, M., Bagheri, E., Lu, W., Ghorbani, A.A.: A detailed analysis of the KDD CUP 99 data set. In: IEEE Symposium on Computational Intelligence for Security and Defense Applications, CISDA 2009, 8–10 July 2009, Ottawa, Canada, pp. 1–6. IEEE (2009)
Stephens, M.: Bayesian analysis of mixture models with an unknown number of components - an alternative to reversible jump methods. Ann. Stat. 40–74 (2000)
Puttini, R.S., Marrakchi, Z., Mé, L.: A Bayesian classification model for real-time intrusion detection. In: AIP Conference Proceedings, vol. 659, pp. 150–162 (2003)
Zhong, S., Khoshgoftaar, T.M., Seliya, N.: Clustering-based network intrusion detection. Int. J. Reliab. Qual. Saf. Eng. 14(02), 169–187 (2007)
Fan, W., Bouguila, N., Ziou, D.: Unsupervised anomaly intrusion detection via localized Bayesian feature selection. In: Cook, D.J., Pei, J., Wang, W., Zaïane, O.R., Wu, X. eds.: 11th IEEE International Conference on Data Mining, ICDM 2011, 11–14 December, Vancouver, BC, Canada, pp. 1032–1037. IEEE Computer Society (2011)
Hartigan, J.A., Wong, M.A.: Algorithm as 136: a k-means clustering algorithm. J. Roy. Stat. Soc. Ser. C (Appl. Stat.) 28(1), 100–108 (1979)
Bouguila, N.: Spatial color image databases summarization. In: Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2007, 15–20 April 2007, Honolulu, Hawaii, USA, pp. 953–956. IEEE (2007)
Bouguila, N., Ziou, D.: Improving content based image retrieval systems using finite multinomial Dirichlet mixture. In: Proceedings of the 2004 14th IEEE Signal Processing Society Workshop Machine Learning for Signal Processing, pp. 23–32. IEEE (2004)
Boutemedjet, S., Ziou, D., Bouguila, N.: Unsupervised feature selection for accurate recommendation of high-dimensional image data. In: Platt, J.C., Koller, D., Singer, Y., Roweis, S.T. (eds.) Advances in Neural Information Processing Systems 20, Proceedings of the Twenty-First Annual Conference on Neural Information Processing Systems, 3–6 December 2007, Vancouver, British Columbia, Canada, pp. 177–184. Curran Associates, Inc. (2007)
Acknowledgment
The completion of this research work was made possible thanks to Concordia University via a Concordia University Research Chair Tier II.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Appendix A
Appendix A
A.1 Derivation of Acceptance Ratio r by Eq. (12)
The derivation of acceptance ratio r is based on the assumption that mixture parameters are independent from each other which means that:
in Eq. (14), since the mixture weigh p is generated following Gibbs sampling method whose acceptance ratio is always 1, it should be excluded from Metropolis-Hastings estimation step. Accordingly, apply the same rule to the proposal distribution as well:
by combining Eqs. (2), (4), (9), (10), (11), (14) and (15), Eq. (12) can be written as follows:
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Fu, S., Bouguila, N. (2018). Bayesian Learning of Finite Asymmetric Gaussian Mixtures. In: Mouhoub, M., Sadaoui, S., Ait Mohamed, O., Ali, M. (eds) Recent Trends and Future Technology in Applied Intelligence. IEA/AIE 2018. Lecture Notes in Computer Science(), vol 10868. Springer, Cham. https://doi.org/10.1007/978-3-319-92058-0_34
Download citation
DOI: https://doi.org/10.1007/978-3-319-92058-0_34
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-92057-3
Online ISBN: 978-3-319-92058-0
eBook Packages: Computer ScienceComputer Science (R0)