Improve Tree Kernel-based Event Pronoun Resolution with Competitive Information
Fang Kong and Guodong Zhou
Event anaphora resolution plays a critical role in discourse analysis. This paper proposes a tree kernel-based framework to event pronoun resolution, the most important type of event anaphora resolution. In particular, a new tree expansion scheme is introduced to automatically determine a proper parse tree structure for event pronoun resolution by considering various kinds of competitive information related with the anaphor and the antecedent candidate. Evaluation on the Onto Notes English corpus shows the appropriateness of the tree kernel-based framework and the effectiveness of competitive information for event pronoun resolution.