Exact Reasoning Under Uncertainty

This paper discusses the use of uncertain knowledge in eipert systems. The discussion is focused on the use of these systems to support decision making in complei, uncertain environments. Within this conteit, our research indicates that strictly probabilistic representations, which enable the use or decision-theoretic reasoning, are highly preferable to recently proposed alternatives. Furthermore, the language of influence diagrams and a corresponding methodology that allows decision theory to be used effectively and efficiently as a decision-making aid is discussed. A sy�tem, called RACHEL, that helps infertile couples select pertinent medical treatments provides an illustration of this methodology.