Extended Bayesian Regression Models: A Symbiotic Application of Belief Networks and Multilayer Perceptrons for the Classification of Ovarian Tumors
We describe a methodology based on a dual Belief Network-Multilayer Perceptron representation to build Bayesian classifiers. This methodology combines efficiently the prior domain knowledge and statistical data. We overview how this Bayesian methodology is able (1) to define constructively a valuable “informative” prior for black-box models, (2) to provide uncertainty information with predictions and (3) to handle missing values based on an auxiliary domain model. We assume that the prior domain model is formalized as a Belief Network (since this representation is a practical solution to acquiring prior domain knowledge) while we use black-box models (such as Multilayer Perceptrons) for learning to utilize the statistical data. In a medical task of predicting the malignancy of ovarian masses we demonstrate these two symbiotic applications of Belief Network models and summarize the practical advantages of the Bayesian approach.
KeywordsBayesian Network Prior Distribution Receiver Operator Characteristic Curve Class Probability Informative Prior Distribution
Unable to display preview. Download preview PDF.
- 1.P. Antal, G. Fannes, S. Van Huffel, B. De Moor, J. Vandewalle, and Dirk Timmerman, Bayesian predictive models for ovarian cancer classiffication: evaluation of logistic regresseion, multi-layer perceptron and belief network models in the bayesian context, Proceedings of the Tenth Belgian-Dutch Conference on Machine Learning, BENELEARN 2000, 2000, pp. 125–132.Google Scholar
- 2.P. Antal, G. Fannes, H. Verrelst, B. De Moor, and J. Vandewalle, Incorporation of prior knowledge in black-box models: Comparison of transformation methods from bayesian network to multilayer perceptrons, Workshop on Fusion of Domain Knowledge with Data for Decision Support, 16th UAI Conf., 2000, pp. 42–48.Google Scholar
- 3.P. Antal, H. Verrelst, D. Timmerman, Y. Moreau, S. Van Huffel, B. De Moor, and I. Vergote, Bayesian networks in ovarian cancer diagnosis: Potential and limitations, Proc. of the 13th IEEE Symp. on Comp.-Based Med.Sys., 2000, Houston, pp. 103–109.Google Scholar
- 4.E. Castillo, J.M. Guttiérrez, and A.S. Hadi, Expert systems and probabilistic network models, Springer, 1997.Google Scholar
- 5.J. Cloete and J.M. Zurada, Knowledge-based neurocomputing, MIT Press, 2000.Google Scholar
- 8.J.A. Hanley et al., The meaning and use of the area under receiver operating characteristic (roc) curve, Radiology 143 (1982), 29–36.Google Scholar
- 9.J.M. Bernardo et al., Bayesian theory, Wiley & Sons, 1995.Google Scholar
- 10.A. Gelman, J.B. Carlin, H.S. Stern., and D.B. Rubin, Bayesian data analysis, Chapman & Hall, 1995.Google Scholar
- 13.R.M. Neal, Bayesian learning for neural networks, Springer, 1996.Google Scholar