Discovery of logic relations for text mining adaptation using unlabeled data

Abstract

We have developed a framework to discover new logic formulae for the adaptation of first-order logic knowledge from a source domain to a target domain for solving the same task. We investigate an approach of adapting the source domain model, represented in Markov Logic Network (MLN), to the target domain using unlabeled data only. The existing logic formulae in the source domain MLN may not be sufficient for the target domain. New logic formulae for are discovered by analyzing the correlations between the candidate relations and the core relations.