Qualitative probabilistic networks with rough-set-based weights

A qualitative probabilistic network (QPN) is the qualitative abstraction of a Bayesian network by encoding variables and the qualitative influences between them in a directed acyclic graph. In order to provide for measuring the weights of qualitative influences and resolving trade-offs during inferences, in this paper we introduce rough-set-based weights to the qualitative influences of QPNs. Looking upon each variable as an equivalence relation on the given sample data table, we give the method to obtain the weights based on the concept of dependency degree in the rough set theory, and learn the enhanced QPN with weighted influences, called EQPN. Then we discuss the conflict-free EQPN inferences and give the method to resolve trade-offs by addressing the symmetry, transitivity and composition properties.