UNCERTAINTY AND PROBABILISTIC REASONING UNCERTAINTY AND PROBABILISTIC REASONING PROBABILISTIC REASONING Context-specific Sign-propagation in Qualitative Probabilistic Networks

Qualitative probabilistic networks represent prob-abilistic influences between variables. Due to the level of representation detail provided, knowledge about influences that hold only in specific contexts cannot be expressed. The results computed from a qualitative network, as a consequence, can be quite weak and uninformative. We extend the ba-sic formalism of qualitative probabilistic networks by providing for the inclusion of context-specific information about influences and show that exploit-ing this information upon inference has the ability to forestall unnecessarily weak results.