Knowledge Structures and Evidential Reasoning in Decision Analysis
This work addresses decision analysis for complex subjects, offering a potentially improved method over existing techniques, though it appears incremental in nature.
The paper tackles the problem of aggregating evidence in complex decision-making by evaluating partially matching factors and applying computational rules, which supports deeper causality expression and better preserves the cognitive structure of decision makers compared to traditional weighting or probabilistic models.
The roles played by decision factors in making complex subject are decisions are characterized by how these factors affect the overall decision. Evidence that partially matches a factor is evaluated, and then effective computational rules are applied to these roles to form an appropriate aggregation of the evidence. The use of this technique supports the expression of deeper levels of causality, and may also preserve the cognitive structure of the decision maker better than the usual weighting methods, certainty-factor or other probabilistic models can.