A probabilistic reasoning system based on a causal graph approach

A description is given of the design and implementation of a probabilistic reasoning system based on a causal graph approach and its application in the field of forensic science. An influence diagram (or a causal graph) is a directed acyclic graph where nodes represent variables and directed links represent probabilistic dependence among variables. For topmost nodes, prior probabilities are usually given. Other nodes are associated with conditional probabilities. The causal graph and corresponding prior and conditional probabilities form the knowledge base to perform reasoning. The probabilistic reasoning system has two distinct stages; the first is to develop a knowledge elicitation procedure through which an expert may supply information about some area of knowledge in the form of a causal graph and associated conditional probabilities; the second stage involves the development of an evidence propagation procedures which will deal with any individual case relating to the area of knowledge which has been dealt with by the first stage. This system would be able to conclude from the current state of the system the probability of any condition or group of conditions occurring