Standard Article

Bayesian Belief Networks

  1. Eric Neufeld,
  2. Michael Horsch

Published Online: 16 MAR 2009

DOI: 10.1002/9780470050118.ecse038

Wiley Encyclopedia of Computer Science and Engineering

Wiley Encyclopedia of Computer Science and Engineering

How to Cite

Neufeld, E. and Horsch, M. 2009. Bayesian Belief Networks. Wiley Encyclopedia of Computer Science and Engineering. 289–298.

Author Information

  1. University of Saskatchewan, Saskatoon, Saskatchewan, Canada

Publication History

  1. Published Online: 16 MAR 2009

Abstract

Bayesian belief networks (BBNs) originated in the 1980s during the course of a debate within the artificial intelligence community about the foundations of uncertain inference. BBNs efficiently store joint probability distributions by exploiting independence relations among variables. The same independencies make inference algorithms more efficient. As well, if the data are rich enough, it is possible to infer the structure of a BBN from raw data, which make BBNs useful for machine learning as well. The structure of BBNs has been given deep causal semantics, which makes BBNs useful for predicting the consequences of interventions.

Keywords:

  • Bayesian belief networks;
  • probabilistic reasoning;
  • uncertainty;
  • causality