Learning Bayesian Network Parameters with Prior Knowledge about Context-Specific Qualitative Influences

We present a method for learning the parameters of a Bayesian network with prior knowledge about the signs of influences between variables. Our method accommodates not just the standard signs, but provides for context-specific signs as well. We show how the various signs translate into order constraints on the network parameters and how isotonic regression can be used to compute order-constrained estimates from the available data. Our experimental results show that taking prior knowledge about the signs of influences into account leads to an improved fit of the true distribution, especially when only a small sample of data is available. Moreover, the computed estimates are guaranteed to be consistent with the specified signs, thereby resulting in a network that is more likely to be accepted by experts in its domain of application.

[1]  Simon Parsons,et al.  Context-specific sign-propagation in qualitative probabilistic networks , 2001, Artif. Intell..

[2]  H. D. Brunk Maximum Likelihood Estimates of Monotone Parameters , 1955 .

[3]  Linda C. van der Gaag,et al.  Learning Bayesian network parameters under order constraints , 2006, Int. J. Approx. Reason..

[4]  Linda C. van der Gaag,et al.  Designing a Procedure for the Acquisition of Probability Constraints for Bayesian Networks , 2004, EKAW.

[5]  Richard E. Neapolitan,et al.  Learning Bayesian networks , 2007, KDD '07.

[6]  David Maxwell Chickering,et al.  Learning Bayesian Networks: The Combination of Knowledge and Statistical Data , 1994, Machine Learning.

[7]  Linda C. van der Gaag,et al.  Defining Classes of Influences for the Acquisition of Probability Constraints for Bayesian Networks , 2004, ECAI.

[8]  J. M. Bremner,et al.  Statistical Inference under Restrictions , 1973 .

[9]  Michael P. Wellman Fundamental Concepts of Qualitative Probabilistic Networks , 1990, Artif. Intell..

[10]  H. D. Brunk Conditional Expectation Given A $\sigma$-Lattice and Applications , 1965 .

[11]  F. T. Wright,et al.  Order restricted statistical inference , 1988 .

[12]  Gregory F. Cooper,et al.  NESTOR: A Computer-Based Medical Diagnostic Aid That Integrates Causal and Probabilistic Knowledge. , 1984 .

[13]  Craig Boutilier,et al.  Context-Specific Independence in Bayesian Networks , 1996, UAI.

[14]  Marek J. Druzdzel,et al.  Building Probabilistic Networks: "Where Do the Numbers Come From?" Guest Editors Introduction , 2000, IEEE Trans. Knowl. Data Eng..

[15]  H. D. Brunk On the Estimation of Parameters Restricted by Inequalities , 1958 .

[16]  Marek J. Druzdzel,et al.  Probabilistic reasoning in decision support systems: from computation to common sense , 1993 .