Abstract
Bayesian networks are models for uncertain reasoning which are achieving a growing importance also for the data mining task of classification. Credal networks extend Bayesian nets to sets of distributions, or credal sets. This paper extends a state-of-the-art Bayesian net for classification, called tree-augmented naive Bayes classifier, to credal sets originated from probability intervals. This extension is a basis to address the fundamental problem of prior ignorance about the distribution that generates the data, which is a commonplace in data mining applications. This issue is often neglected, but addressing it properly is a key to ultimately draw reliable conclusions from the inferred models. In this paper we formalize the new model, develop an exact linear-time classification algorithm, and evaluate the credal net-based classifier on a number of real data sets. The empirical analysis shows that the new classifier is good and reliable, and raises a problem of excessive caution that is discussed in the paper. Overall, given the favorable trade-off between expressiveness and efficient computation, the newly proposed classifier appears to be a good candidate for the wide-scale application of reliable classifiers based on credal networks, to real and complex tasks.
Similar content being viewed by others
References
Abell'an, J. and Moral, S.: Building Classification Trees Using the Total Uncertainty Criterion, in: deCooman, G., Fine, T., and Seidenfeld, T. (eds), ISIPTA'01, Shaker Publishing, TheNetherlands, 2001, pp. 1–8.
Balas, E. and Zemel, E.: An Algorithm for Large Zero-One Knapsack Problems, Operations Research 28 (1980), pp. 1130–1154.
Bernard, J.-M.: Implicative Analysis for Multivariate Binary Data Using an Imprecise Dirichlet Model, Journal of Statistical Planning and Inference 105 (1) (2002), pp. 83–103.
Bernardo, J.M. and Smith, A. F. M.: Bayesian Theory, Wiley, New York, 1996.
Campos, L., Huete, J., and Moral, S.: Probability Intervals: A Tool for Uncertain Reasoning, International Journal of Uncertainty, Fuzziness, and Knowledge-Based Systems 2(2) (1994), pp. 167–196.
Chen, T. T. and Fienberg, S. E.: Two-Dimensional Contingency Tables with both Completely and Partially Cross-Classifled Data, Biometrics 32 (1974), pp. 133–144.
Chow, C. K. and Liu, C. N.: Approximating Discrete Probability Distributions with Dependence Trees, IEEE Transactions on Information Theory IT-14(3) (1968), pp. 462–467.
Couso, I., Moral, S., and Walley, P.: A Survey of Concepts of Independence for Imprecise Probability, Risk, Decision and Policy 5 (2000), pp. 165–181.
Cozman, F. G.: Credal Networks, Artificial Intelligence 120 (2000), pp. 199–233.
Cozman, F.G.: Separation Properties of Sets of Probabilities, in: Boutilier, C. and Goldszmidt, M. (eds), UAI-2000, Morgan Kaufmann, San Francisco, 2000, pp. 107–115.
Dasgupta, S.: Learning Polytrees, in: UAI-99,Morgan Kaufmann, San Francisco, 1999, pp. 134– 141.
Duda, R. O. and Hart, P. E.: Pattern Classification and Scene Analysis, Wiley, New York, 1973.
Duda, R. O., Hart, P. E., and Stork, D. G.: Pattern Classification, 2nd edition, Wiley, 2001.
Fagiuoli, E. and Zaffalon, M.: 2U: An Exact Interval Propagation Algorithm for Polytrees with Binary Variables, Artificial Intelligence 106(1) (1998), pp. 77–107.
Fagiuoli, E. and Zaffalon, M.: Tree-Augmented Naive Credal Classifiers, in: IPMU 2000: Proceedings of the 8th Information Processing andManagement of Uncertainty in Knowledge-Based Systems Conference, Universidad Polit'ecnica de Madrid, Spain, 2000, pp. 1320–1327.
Fayyad, U. M. and Irani, K. B.: Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning, in: Proceedings of the 13th International Joint Conference on Artificial Intelligence, Morgan Kaufmann, San Francisco, 1993, pp. 1022–1027.
Ferreira da Rocha, J. C. and Cozman, F. G.: Inference with Separately Specified Sets of Probabilities in Credal Networks, in: Darwiche, A. and Friedman, N. (eds), Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence (UAI-2002), Morgan Kaufmann, 2002, pp. 430–437.
Friedman, N., Geiger, D., and Goldszmidt, M.: Bayesian Networks Classifiers,Machine Learning 29 (2/3) (1997), pp. 131–163.
Ha, V., Doan, A., Vu, V., and Haddawy, P.: Geometric Foundations for Interval-Based Probabilities, Annals of Mathematics and Artificial Intelligence 24(1–4) (1998), pp. 1–21.
Kleiter, G. D.: The Posterior Probability of BayesNets with Strong Dependences, Soft Computing 3 (1999), pp. 162–173.
Kohavi, R.: A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection, in: IJCAI-95, Morgan Kaufmann, San Mateo, 1995, pp. 1137–1143.
Kullback, S. and Leiber, R. A.: On Information and Sufficiency, Ann. Math. Statistics 22 (1951), pp. 79–86.
Kyburg, H. E. Jr.: Rational Belief, The Behavioral and Brain Sciences 6 (1983), pp. 231–273.
Lawler, E.: Fast Approximation Algorithms for Knapsack Problems, Mathematics of Operations Research 4(4) (1979), pp. 339–356.
Levi, I.: The Enterprise of Knowledge, MIT Press, London, 1980.
Little, R. J. A. and Rubin, D. B.: Statistical Analysis with Missing Data,Wiley, New York, 1987.
Martello, S. and Toth, P.: Knapsack Problems: Algorithms andComputer Implementations,Wiley, Chichester, 1990.
Moral, S. and Cano, A.: Strong Conditional Independence for Credal Sets, Annals ofMathematics and Artificial Intelligence 35(1–4) (2002), pp. 295–321.
Murphy, P. M. and Aha, D. W.: UCI Repository of Machine Learning Databases, 1995, http://www.sgi.com/Technology/mlc/db/.
Nivlet, P., Fournier, F., and Royer, J.-J.: Interval Discriminant Analysis: An Efficient Method to Integrate Errors in Supervised PatternRecognition, in: de Cooman, G., Fine, T., and Seidenfeld, T. (eds), ISIPTA'01, Shaker Publishing, The Netherlands, 2001, pp. 284–292.
Pearl, J.: Probabilistic Reasoning in Intelligent Systems:Networks of Plausible Inference,Morgan Kaufmann, San Mateo, 1988.
Quinlan, J. R.: C4.5: Programs for Machine Learning, Morgan Kaufmann, San Mateo, 1993.
Ramoni, M. and Sebastiani, P.: Robust Bayes Classifiers, Artificial Intelligence 125(1–2) (2001), pp. 209–226.
Walley, P.: Inferences from Multinomial Data: Learning about a Bag of Marbles, J. R. Statist. Soc. B 58(1) (1996), pp. 3–57.
Walley, P.: Statistical Reasoning with Imprecise Probabilities, Chapman and Hall, New York, 1991.
Walley, P. and Fine, T. L.: Towards a Frequentist Theory of Upper and Lower Probability, Ann. Statist. 10 (1982), pp. 741–761.
Witten, I. H. and Frank, E.: Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations, Morgan Kaufmann, 1999.
Zaffalon, M.: A Credal Approach to Naive Classification, in: de Cooman, G., Cozman, F., Moral, S., and Walley, P. (eds), ISIPTA'99, The Imprecise Probabilities Project, Univ. of Gent, Belgium, 1999, pp. 405–414.
Zaffalon, M.: Statistical Inference of the Naive Credal Classifier, in: de Cooman, G., Fine, T., and Seidenfeld, T. (eds), ISIPTA'01, Shaker Publishing, The Netherlands, 2001, pp. 384–393.
Zaffalon, M.: The Naive Credal Classifier, Journal of Statistical Planning and Inference 105(1) (2002), pp. 5–21.
Zaffalon, M. and Hutter, M.: Robust Feature Selection by Mutual Information Distributions, in: Darwiche, A. and Friedman, N. (eds), Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence (UAI-2002), Morgan Kaufmann, 2002, pp. 577–584.
Zaffalon, M. and Hutter, M.: Robust Inference of Trees, Technical Report IDSIA–11–03, IDSIA, 2003.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Zaffalon, M., Fagiuoli, E. Tree-Based Credal Networks for Classification. Reliable Computing 9, 487–509 (2003). https://doi.org/10.1023/A:1025822321743
Issue Date:
DOI: https://doi.org/10.1023/A:1025822321743