Abstract
This paper addresses classification problems in which the class membership of training data is only partially known. Each learning sample is assumed to consist in a feature vector and an imprecise and/or uncertain “soft” label m i defined as a Dempster-Shafer basic belief assignment over the set of classes. This framework thus generalizes many kinds of learning problems including supervised, unsupervised and semi-supervised learning. Here, it is assumed that the feature vectors are generated from a mixture model. Using the General Bayesian Theorem, we derive a criterion generalizing the likelihood function. A variant of the EM algorithm dedicated to the optimization of this criterion is proposed, allowing us to compute estimates of model parameters. Experimental results demonstrate the ability of this approach to exploit partial information about class labels.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Ambroise, C., Govaert, G.: EM algorithm for partially known labels. In: Proc. of IFCS 2000 (Namur, Belgium), pp. 161–166. Springer, Heidelberg (2000)
Cobb, B.R., Shenoy, P.P.: On the plausibility transformation method for translating belief function models to probability models. Internat. J. Approx. Reason 41(3), 314–330 (2006)
Dempster, A.P.: Upper and lower probabilities induced by a multivalued mapping. Ann. Math. Statist. 38, 325–339 (1967)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B. Stat. Methodol. 39, 1–38 (1977)
Denœux, T.: A k-nearest neighbor classification rule based on Dempster-Shafer theory. IEEE Trans. Systems Man Cybernet 25(05), 804–813 (1995)
Denœux, T., Zouhal, L.M.: Handling possibilistic labels in pattern classification using evidential reasoning. Fuzzy Sets Syst. 122(3), 47–62 (2001)
Dubois, D., Prade, H.: On the unicity of Dempster’s rule of combination. Int. J. Intell. Syst. 1, 133–142 (1986)
Elouedi, Z., Mellouli, K., Smets, P.: Belief decision trees: Theoretical foundations. Internat. J. Approx. Reason 28, 91–124 (2001)
Grandvallet, Y.: Logistic regression for partial labels. In: Proc. of IPMU 2002 (Annecy, France), vol. III, pp. 1935–1941 (2002)
Hosmer, D.W.: A comparison of iterative maximum likelihood estimates of the parameters of a mixture of two normal distributions under three different types of sample. Biometrics 29, 761–770 (1973)
Hüllermeier, E., Beringer, J.: Learning from ambiguously labeled examples. In: Famili, A.F., Kok, J.N., Peña, J.M., Siebes, A., Feelders, A. (eds.) IDA 2005. LNCS, vol. 3646. Springer, Heidelberg (2005)
Jraidi, I., Elouedi, Z.: Belief classification approach based on generalized credal EM. In: Mellouli, K. (ed.) ECSQARU 2007. LNCS (LNAI), vol. 4724, pp. 524–535. Springer, Heidelberg (2007)
Monney, P.-A.: A Mathematical Theory of Arguments for Statistical Evidence. Contributions to Statistics. Physica-Verlag, Heidelberg (2003)
Shafer, G.: A mathematical theory of evidence. Princeton University Press, Princeton (1976)
Shenoy, P.P., Giang, P.H.: Decision making on the sole basis of statistical likelihood. Artificial Intelligent 165(2), 137–163 (2005)
Smets, P.: Possibilistic inference from statistical data. In: Proc. of WCMSM 1982 (Las-Palmas, Spain), pp. 611–613 (1982)
Smets, P.: Belief functions: the disjunctive rule of combination and the generalized Bayesian theorem. Internat. J. Approx. Reason 9, 1–35 (1993)
Smets, P.: Quantifying beliefs by belief functions: An axiomatic justification. In: Proc. of IJCAI 1993 (Chambéry), vol. 1, pp. 598–603 (1993)
Smets, P.: Belief functions on real numbers. Internat. J. Approx. Reason 40(3), 181–223 (2005)
Smets, P., Kennes, R.: The Transferable Belief Model. Artificial Intelligence 66, 191–243 (1994)
Vannoorenberghe, P., Smets, P.: Partially supervised learning by a credal EM approach. In: Godo, L. (ed.) ECSQARU 2005. LNCS (LNAI), vol. 3571, pp. 956–967. Springer, Heidelberg (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Côme, E., Oukhellou, L., Denœux, T., Aknin, P. (2008). Mixture Model Estimation with Soft Labels. In: Dubois, D., Lubiano, M.A., Prade, H., Gil, M.Á., Grzegorzewski, P., Hryniewicz, O. (eds) Soft Methods for Handling Variability and Imprecision. Advances in Soft Computing, vol 48. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-85027-4_21
Download citation
DOI: https://doi.org/10.1007/978-3-540-85027-4_21
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-85026-7
Online ISBN: 978-3-540-85027-4
eBook Packages: EngineeringEngineering (R0)