Learning probabilistic Description logic concepts: under different Assumptions on missing knowledge

Abstract

Knowledge available through Semantic Web standards can be missing, generally because of the adoption of the Open World Assumption. We present a Statistical Relational Learning system for learning terminological naïve Bayesian classifiers, which estimate the probability that an individual belongs to a target concept given its membership to a set of Description Logic concepts. During the learning process, we consistently handle the lack of knowledge that may be introduced by the adoption of the Open World Assumption, depending on the varying nature of the missing knowledge itself.