Article
Full entry |
PDF
(0.3 MB)
Feedback
PDF
(0.3 MB)
Feedback
Keywords:
Bayesian networks; noisy-or model; classification; generalized linear models
Bayesian networks; noisy-or model; classification; generalized linear models
Summary:
In this paper, we generalize the noisy-or model. The generalizations are three-fold. First, we allow parents to be multivalued ordinal variables. Second, parents can have both positive and negative influences on their common child. Third, we describe how the suggested generalization can be extended to multivalued child variables. The major advantage of our generalizations is that they require only one parameter per parent. We suggest a model learning method and report results of experiments on the Reuters text classification data. The generalized noisy-or models achieve equal or better performance than the standard noisy-or. An important property of the noisy-or model and of its generalizations suggested in this paper is that it allows more efficient exact inference than logistic regression models do.
References:
[1] Almond, R. G., Mislevy, R. J., Steinberg, L., Yan, D., Williamson, D.: Bayesian Networks in Educational Assessment. Statistics for Social and Behavioral Sciences. Springer, New York 2015. DOI 10.1007/978-1-4939-2125-6 | MR 3328529
[2] Apt{é}, Ch., Damerau, F., Weiss, S. M.: Automated learning of decision rules for text categorization. ACM Trans. Inform. Syst. 12 (1994), 3, 233-251. DOI 10.1145/183422.183423
[3] Byrd, R. H., Lu, P., Nocedal, J., Zhu, C.: A limited memory algorithm for bound constrained optimization. SIAM J. Sci. Comput. 16 (1995), 1190-1208. DOI 10.1137/0916069 | MR 1346301 | Zbl 0836.65080
[4] Díez, F. J.: Parameter adjustment in Bayes networks. The generalized noisy OR gate. In: Proc. Ninth Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann 1993, pp. 99-105. DOI 10.1016/b978-1-4832-1451-1.50016-0
[5] Díez, F. J., Druzdzel, M. J.: Canonical Probabilistic Models for Knowledge Engineering. Technical Report CISIAD-06-01, UNED, Madrid 2006.
[6] Díez, F. J., Galán, S. F.: An efficient factorization for the noisy MAX. Int. J. Intell. Syst. 18 (2003), 165-177. DOI 10.1002/int.10080
[7] Heckerman, D., Breese, J.: A new look at causal independence. In: Proc. Tenth Conference on Uncertainty in Artificial Intelligence, Seattle, Morgan Kaufmann 1994, pp. 286-292. DOI 10.1016/b978-1-55860-332-5.50041-9
[8] Henrion, M.: Practical issues in constructing a Bayes' Belief Network. In: Proc. Third Conference Annual Conference on Uncertainty in Artificial Intelligence, AUAI Press 1987, pp. 132-139.
[9] Jensen, F. V., Nielsen, T. D.: Bayesian Networks and Decision Graphs. Second edition. Springer, 2007. DOI 10.1007/978-0-387-68282-2 | MR 2344166
[10] McCullagh, P.: Regression models for ordinal data. J. Roy. Statist. Soc. Series B (Methodological) 42 (1980), 109-142. MR 0583347 | Zbl 0483.62056
[11] McCullagh, P., Nelder, J. A.: Generalized Linear Models. Chapman and Hall, London 1989. DOI 10.1007/978-1-4899-3242-6 | MR 3223057 | Zbl 0744.62098
[12] Miller, R. A., Fasarie, F. E., Myers, J. D.: Quick medical reference (QMR) for diagnostic assistance. Medical Comput. 3 (1986), 34-48.
[13] Neal, R. M.: Connectionist learning of belief networks. Artif. Intell. 56 (1992), 1, 71-113. DOI 10.1016/0004-3702(92)90065-6 | MR 1171969 | Zbl 0761.68081
[14] Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufman, San Mateo 1988. MR 0965765 | Zbl 0746.68089
[15] Team, R Development Core: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna 2008.
[16] Rijmen, F.: Bayesian networks with a logistic regression model for the conditional probabilities. Int. J. Approx. Reas. 48 (2008), 2, 659-666. DOI 10.1016/j.ijar.2008.01.001 | MR 2419024 | Zbl 1184.62039
[17] Samejima, F.: Estimation of Latent Ability Using a Response Pattern of Raded Scores (Psychometric Monograph No. 17). Psychometric Society, Richmond 1969.
[18] Saul, L. K., Jaakkola, T., Jordan, M. I.: Mean field theory for sigmoid belief networks. J. Artif. Intell. Res. 4 (1996), 61-76. Zbl 0900.68379
[19] Savický, P., Vomlel, J.: Exploiting tensor rank-one decomposition in probabilistic inference. Kybernetika 43 (2007), 5, 747-764. MR 2376335 | Zbl 1148.68539
[20] Srinivas, S.: A generalization of the noisy-or model. In: Proc. Ninth Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann 1993, pp. 208-215. DOI 10.1016/b978-1-4832-1451-1.50030-5
[21] Vomlel, J.: Noisy-or classifier. Int. J. Intell. Syst. 21 (2006), 381-398. DOI 10.1002/int.20141 | Zbl 1160.68584
[22] Vomlel, J.: A generalization of the noisy-or model to multivalued parent variables. In: Proc. 16th Czech-Japan Seminar on Data Analysis and Decision Making under Uncertainty 2013, pp. 19-27.
[23] Vomlel, J., Tichavský, P.: On tensor rank of conditional probability tables in Bayesian networks. A preprint arXiv:1409.6287, 2014. MR 3178417
[24] Zagorecki, A., Druzdzel, M. J.: Knowledge engineering for Bayesian networks: How common are noisy-MAX distributions in practice?. IEEE Trans. Systems, Man, and Cybernetics: Systems 43 (2013) 186-195. DOI 10.1109/tsmca.2012.2189880
Search
Partner of
