Results 1 
3 of
3
Accelerated Quantification of Bayesian Networks with Incomplete Data
 In Proceedings of First International Conference on Knowledge Discovery and Data Mining
, 1995
"... Probabilistic expert systems based on Bayesian networks (BNs) require initial specification of both a qualitative graphical structure and quantitative assessment of conditional probability tables. This paper considers statistical batch learning of the probability tables on the basis of incomple ..."
Abstract

Cited by 28 (2 self)
 Add to MetaCart
Probabilistic expert systems based on Bayesian networks (BNs) require initial specification of both a qualitative graphical structure and quantitative assessment of conditional probability tables. This paper considers statistical batch learning of the probability tables on the basis of incomplete data and expert knowledge. The EM algorithm with a generalized conjugate gradient acceleration method has been dedicated to quantification of BNs by maximum posterior likelihood estimation for a superclass of the recursive graphical models. This new class of models allows a great variety of local functional restrictions to be imposed on the statistical model, which hereby extents the control and applicability of the constructed method for quantifying BNs. Introduction The construction of probabilistic expert systems (Pearl 1988, Andreassen et al. 1989) based on Bayesian networks (BNs) is often a challenging process. It is typically divided into two parts: First the constructi...
Bayesian Applications of Belief Networks and Multilayer Perceptrons for Ovarian Tumor Classification with Rejection
, 2003
"... Incorporating prior knowledge into blackbox classifiers is still much of an open problem. We propose a hybrid Bayesian methodology that consists in encoding prior knowledge in the form of a (Bayesian) belief network and then using this knowledge to estimate an informative prior for a blackbox model ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
Incorporating prior knowledge into blackbox classifiers is still much of an open problem. We propose a hybrid Bayesian methodology that consists in encoding prior knowledge in the form of a (Bayesian) belief network and then using this knowledge to estimate an informative prior for a blackbox model (e.g. a multilayer perceptron). Two technical approaches are proposed for the transformation of the belief network into an informative prior. The first one consists in generating samples according to the most probable parameterization of the Bayesian belief network and using them as virtual data together with the real data in the Bayesian learning of a multilayer perceptron. The second approach consists in transforming probability distributions over belief network parameters into distributions over multilayer perceptron parameters. The essential attribute of the hybrid methodology is that it combines prior knowledge and statistical data efficiently when prior knowledge is available and the sample is of small or medium size. Additionally, we describe how the Bayesian approach can provide uncertainty information about the predictions (e.g. for classification with rejection). We demonstrate these techniques on the medical task of predicting the malignancy of ovarian masses and summarize the practical advantages of the Bayesian approach. We compare the learning curves for the hybrid methodology with those of several belief networks and multilayer perceptrons. Furthermore, we report the performance of Bayesian belief networks when they are allowed to exclude hard cases based on various measures of prediction uncertainty.
Integrated Approach to Total Maximum Daily Load Development for Neuse River Estuary using Bayesian Probability Network Model „NeuBERN…
"... Abstract: We develop a probability network model to characterize eutrophication in the Neuse River Estuary, North Carolina, and support the estimation of a total maximum daily load �TMDL � for nitrogen. Unlike conventional simulation models, probability networks describe probabilistic dependencies a ..."
Abstract
 Add to MetaCart
Abstract: We develop a probability network model to characterize eutrophication in the Neuse River Estuary, North Carolina, and support the estimation of a total maximum daily load �TMDL � for nitrogen. Unlike conventional simulation models, probability networks describe probabilistic dependencies among system variables rather than substance mass balances. Full networks are decomposable into smaller submodels, with structure and quantification that reflect relevant theory, judgment, and/or observation. Model predictions are expressed probabilistically, which supports consideration of frequencybased water quality standards and explicit estimation of the TMDL margin of safety. For the Neuse Estuary TMDL application, the probability network can be used to predict compliance with the dissolved oxygen and chlorophyll a regulatory criteria as a function of riverine nitrogen load. In addition, the model includes ecological endpoints, such as fishkills and shellfish survival, that are typically more meaningful to stakeholders than conventional water quality characteristics. Incorporating these unregulated attributes into TMDL decisions will require explicit consideration of costs, benefits, and relative likelihoods of various possible outcomes under alternate loading scenarios.