Proceedings Abstracts of the Twenty-Third International Joint Conference on Artificial Intelligence

Joint Modeling of Argument Identification and Role Determination in Chinese Event Extraction with Discourse-Level Information / 2120
Peifeng Li, Qiaoming Zhu, Guodong Zhou

Argument extraction is a challenging task in event extraction. However, most of previous studies focused on intra-sentence information and failed to extract inter-sentence arguments. This paper proposes a discourse-level joint model of argument identification and role determination to infer those inter-sentence arguments in a discourse. Moreover, to better represent the relationship among relevant event mentions and the relationship between an event mention and its arguments in a discourse, this paper introduces various kinds of corpus-based and discourse-based constraints in the joint model, either automatically learned or linguistically motivated. Evaluation on the ACE 2005 Chinese corpus justifies the effectiveness of our joint model over a strong baseline in Chinese argument extraction, in particular argument identification.