Refining Reasoning in Qualitative Probabilistic Networks
This work addresses limitations in qualitative probabilistic reasoning for AI and decision-making systems, but it appears incremental as it builds on existing non-numerical approaches.
The paper tackles the problem of qualitative probabilistic networks being unable to predict probability changes or identify the most likely hypothesis due to their simple value sets, and it proposes methods to resolve these issues by refining the representation.
In recent years there has been a spate of papers describing systems for probabilisitic reasoning which do not use numerical probabilities. In some cases the simple set of values used by these systems make it impossible to predict how a probability will change or which hypothesis is most likely given certain evidence. This paper concentrates on such situations, and suggests a number of ways in which they may be resolved by refining the representation.