Qualitative probabilistic networks (QPNs) are a qualitative abstraction of Bayesian networks, which focus on the monotonic relationship between variables. However, sometimes we don’t care about this monotonic relationship, but more con-cerned about the change of probability of the variable comparing with its prior proba-bility. In this paper, we propose a new qualitative abstraction of Bayesian networks—PQPNs (prior qualitative probabilistic networks) that focuses on the prior probability distribution. Our main work is:1. Definition of PQPN. We will analyze the properties of PQPN, namely symmetry, transitivity and composition, and design sign propagation algorithm of PQPN.2. Relationship between QPN and PQPN, which mainly includes relationship of qualitative influence between them and proof of equivalence between them in binary problems.3. Verify the correctness of sign propagation algorithm of PQPN and compare the results of QPNs and PQPNs by experiments. |