ArThUR: A Tool for Markov Logic Network

Abstract

Logical approaches-and ontologies in particular-offer a well-adapted framework for representing knowledge present on the Semantic Web ( Open image in new window ). These ontologies are formulated in Open image in new window ( Open image in new window ), which are based on expressive Open image in new window ( Open image in new window ). Open image in new window are a subset of Open image in new window ( Open image in new window ) that provides decidable reasoning. Based on Open image in new window , it is possible to rely on inference mechanisms to obtain new knowledge from axioms, rules and facts specified in the ontologies. However, these classical inference mechanisms do not deal with : Open image in new window probabilities. Several works recently targeted those issues (i.e. Open image in new window , Open image in new window , Open image in new window , etc.), but none of them combines Open image in new window with Open image in new window ( Open image in new window ) formalism. Several open source software packages for Open image in new window are available (e.g. Open image in new window , Open image in new window , Open image in new window , etc.). In this paper, we present Open image in new window , a Java framework for reasoning with probabilistic information in the Open image in new window . Open image in new window incorporate three open source software packages for Open image in new window , which is able to reason with uncertainty information, showing that it can be used in several real-world domains. We also show several experiments of our tool with different ontologies.