Adaptive Clustering for Coreference Resolution with Deterministic Rules and Web-Based Language Models

We present a novel adaptive clustering model for coreference resolution in which the expert rules of a state of the art deterministic system are used as features over pairs of clusters. A significant advantage of the new approach is that the expert rules can be easily augmented with new semantic features. We demonstrate this advantage by incorporating semantic compatibility features for neutral pronouns computed from web n-gram statistics. Experimental results show that the combination of the new features with the expert rules in the adaptive clustering approach results in an overall performance improvement, and over 5% improvement in F1 measure for the target pronouns when evaluated on the ACE 2004 newswire corpus.