COMBINING OPINIONS FROM SEVERAL EXPERTS

Abstract
This paper addresses the problem of aggregating a number of expert opinions that have been expressed in some numerical form. An important feature of sets of expert opinions is the possibility of stochastic dependence between members of the sets. We develop an approach for combining expert opinions that formally allows for such dependence. This approach is based on an extension of the Dempster-Shafer theory, a well-known calculus for reasoning with uncertainty in artificial intelligence.