An Interactive Program Fo Conversational Elicitation of Decision Structures

Abstract
An interactive computer program has been designed and implemented that elicits a decision tree from a decisionmaker in an English-like conversational mode. It emulates a decision analyst who guides the decisionmaker in structuring and organizing his knowledge about a particular problem domain. The objectives of the research were: 1) to provide the decision analysis industry with a practical automated tool for eliciting decision structures where manual elicitation techniques are either infeasible or uneconomical, 2) to cast the decision analyst's behavior into a formal framework in order to examine the principles governing the elicitation procedure and gain a deeper understanding of the analysis process itself, and 3) to provide experimental psychologists with an automated research tool for coding subjects' perception of problem situations into a standard and formal representation. The approach centers on the realization that the process of conducting an elicitation dialogue is structurally identical to conducting a heuristic search on game trees, as is commonly practiced in artificial intelligence programs. Heuristic search techniques, when applied to tree elicitation, permit real-time rollback and sensitivity analysis as the tree is being formulated. Thus it is possible to concentrate effort on expanding those parts of the tree which are crucial for the resolution of the solution plan. The program requires the decisionmaker to provide provisional values at each intermediate stage in the tree construction, which estimate the promise of future opportunities open to him from that stage.

This publication has 2 references indexed in Scilit: