A continuous possibility propagation diagram approach for reasoning under uncertainty

Reasoning under uncertainty is an important issue in artificial intelligence systems. A dynamic causality trees-diagram based method capable of dealing with complex cases like causality loops has been presented in Qin Zang (1994). But it, like most existing methods, considers only discrete cases and thus restricts its applications. Developed from it, this paper presents a new method to deal with continuous cases in which the ascendant, descendent and linkage variables can be continuous while keeping them independent of each other. Probability theory can not be rigorously applied but is somewhat relaxed. Therefore the uncertainty measure is called possibility instead of probability. An example is given to illustrate the method and show its new features