Learning bayesian-neural network from mixed-mode data

Authors:
Wang Limin
Affiliations:
Key Laboratary of Symbol Computation and Knowledge Engineering of Ministry, of Education, JiLin University, ChangChun, China
Venue:
ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I
Year:
2006

Citing 6
Cited 0

Probabilistic neural networks

Neural Networks
Semi-naive Bayesian classifier

EWSL-91 Proceedings of the European working session on learning on Machine learning
Bayesian Network Classifiers

Machine Learning - Special issue on learning with probabilistic representations
Bayesian Network Classification with Continuous Attributes: Getting the Best of Both Discretization and Parametric Fitting

ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Probabilistic fault localization in communication systems using belief networks

IEEE/ACM Transactions on Networking (TON)
An analysis of Bayesian classifiers

AAAI'92 Proceedings of the tenth national conference on Artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

For reasoning with uncertain knowledge the use of probability theory has been broadly investigated. This paper proposed a novel probabilistic network named Bayesian-Neural Network (BNN). BNN reduces computational complexity by dividing input attribute set into two parts, each modelled by Bayesian network or Neural network. The outputs produced by different classifiers is then solved in the output space by estimating the class-conditional structural mixtures. Empirical studies on a set of natural domains show that BNN has clear advantages with respect to the generalization ability.