Graph-cut-based anaphoricity determination for coreference resolution
Vincent Ng · 2009
Recent work has shown that explicitly identifying and filtering non-anaphoric mentions prior to coreference resolution can improve the performance of a coreference system. We present a novel approach to this task of anaphoricity determination based on graph cuts, and demonstrate its superiority to competing approaches by comparing their effectiveness in improving a learning-based coreference system on the ACE data sets.