Task driven coreference resolution for relation extraction

Abstract

This paper presents the extension of an existing mim-imally supervised rule acquisition method for relation extraction by coreference resolution (CR). To this end, a novel approach to CR was designed and tested. In comparison to state-of-the-art methods for CR, our strategy is driven by the target semantic relation and utilizes domain-specific ontological and lexical knowledge in addition to the learned relation extraction rules. An empirical investigation reveals that newswire texts in our selected domains contain more coreferring noun phrases than prononimal coreferences. This means that existing methods for CR would not suffice and a semantic approach is needed. Our experiments show that the utilization of domain knowledge can boost CR. In our approach, the tasks of relation extraction and CR support each other. On the one hand, reference resolution is needed for the detection of arguments of the target relation. On the other hand, domain modelling for the IE task is used for semantic classification of the referring nouns. Moreover, the application of the learned relation extraction rules often narrows down the number of candidates for CR. With respect to the minimally supervised learning of relation extraction grammars, we design and evaluate two integration strategies: (i) resolution after the complete pattern acquisition process and (ii) resolution embedded in the iterations of the learning process. The evaluation helps us to gain and substantiate a relevant insight: CR effectively improves recall in both strategies but it can hurt the precision because of its error spreading potential.