Back to Methods and Tools
Monday, 25 October 2010
Bayesian Belief Networks
Bayesian Belief Networks (BBN) facilitate the sharing and integration of disparate sources and forms of knowledge. A BBN is essentially a cause and effect diagram (flow diagram) in which nodes present factors believed (by different stakeholders) to influence particular outcomes.
BBN is based on the work of Thomas Bayes, and his theory on conditional probability – Bayes’ Theorem, published posthumously in 1764. (Stanford Encyclopedia of Philosphy 2003)
“…a Bayesian belief network is a model. It can be a model of anything: the weather, a disease and its symptoms, a military battalion, even a garbage disposal. Belief networks are especially useful when the information about the past and/or the current situation is vague, incomplete, conflicting, and uncertain.” (Charles River Analytics 2004:4)
BBN’s are models for representing uncertainty in our knowledge (Greiner, online). The models are represented graphically, normally by a flow chart generated by specifically designed software.
Each variable in the model is represented by a node, and causal relationships are represented by an arrow, called an edge (Charles River Analytics 2004:9)
The following example has been taken from the Norsys Software Corp. Bayes Net Library, which contains useful examples and information about Bayesian Belief Networks.
A very small belief network for a fictitious medical example about whether a patient has tuberculosis, lung cancer or bronchitis, related to their X-ray, dyspnea, visit-to-Asia and smoking status (Norsys Bayes Net Library, online).
- Use specifically designed software, and make sure you know how to use it.
- Keep in mind that there may be computational difficulties in exploring previously unknown networks and beliefs (Niedermayer 1998).
Cain, J. D., C. H. Batchelor, and D. K. N. Waughray. 1999. Belief networks: a framework for the participatory development of natural resource management strategies. Environment,Development and Sustainability 1:123–133.
Charles River Analytics. 2004. About Bayesian Belief Networks
www.cra.com/pdf/BNetBuilderBackground.pdf (accessed 14 September 2009)
Norsys Software Corp. Bayes Net Library, Asia, http://www.norsys.com/netlib/asia.htm (31 October 2009). Examples taken from: Lauritzen, Steffen L. and David J. Spiegelhalter (1988) "Local computations with probabilities on graphical structures and their application to expert systems". Journal of the Royal Statistical Society: Series B, 50(2):157-194.
Stanford Encyclopedia of Philosophy. 2003. Bayes’ Theorem. http://plato.stanford.edu/entries/bayes-theorem/ (accessed 31 October 2009)
Drudzel 1996. Qualitative Verbal Explanations in Bayesian Belief Networks. Artificial Intelligence and Simulation of Behaviour Quarterly,Special issue on Bayesian Belief Networks, 94:43-54, http://www.pitt.edu/~druzdzel/psfiles/aisb.pdf (accessed 14 September 2009)
Greiner, Russell. Bayesian Belief Nets, http://www.cs.ualberta.ca/~greiner/bn.html
Hamilton, G, and C. Alston, T. Chiffings, E. Abal, B. Hart and K. Mengersen. 2005. Integrating Science Through Bayesian Belief Networks: Case Study of Lyngbya in Moreton Bay. In: International Congress on Modelling and Simulation (MODSIM05), 12-15 December 2005, Australia, Victoria, Melbourne. http://www.mssanz.org.au/modsim05/papers/hamilton.pdf
Niedermayer, Daryle. 1998. An Introduction to Bayesian Networks and their Contemporary Applications, http://www.niedermayer.ca/papers/bayesian/bayes.html#table (accessed 31 October 2009)
Marcot, Bruce G. 2009. Creating Bayesian Network Models in Ecology, http://www.aracnet.com/~brucem/bbns.htm (accessed 31 October 2009).
Wooldridge, Scott. 2003. Bayesian Belief Networks. www.mrcmekong.org/download/.../bayesian_network_reading2.pdf, (accessed 14 September 2009)