BELIEF CONDITIONING UNDER VARIOUS EXTERNAL CONDITIONS

Uncertain information processing is one of the important areas of artificial intelligence. Probability theory is a powerful tool to deal with chances of random events. The conceptual basis of the theory is the concept of a complete group of events, where for each event the probability of its occurrence is somehow evaluated. In case if the initial information and/or knowledge is not sufficient, it might be difficult to evaluate all those probabilities. In order to correctly use such limited information and/or the knowledge that cannot be used by probability theory, theory of evidence has been elaborated [1]. In literature, this theory is frequently called Dempster-Shafer theory. G.Shafer has employed Dempster's concept of multivalued probabilistic mappings and has developed an in principle new theory.