In this paper we study the problem of obtaining a correspondence between Bayesian networks and neural networks. It is shown how such a correspondence is established by obtaining a mathematical function which relates the parameters of the two models. We show the validity of our method by deriving the parameters to be used in a Bayesian network constructed to combine GIS data for assessing the risk of desertification of burned forest areas in the Mediterranean region.
- Bayesian networks
- Conditional probability matrices
- Geographic information processing
- Neural networks