Exploiting Role-Identifying Nouns and Expressions for Information Extraction
William Phillips, Ellen Riloff · Recent Advances in Natural Language Processing · 2007
We present a new approach for extraction pattern learning that exploits role-identifying nouns, which are nouns whose semantics reveal the role that they play in an event (e.g., an “assassin” is a perpetrator). Given a few seed nouns, a bootstrapping algorithm automatically learns roleidentifying nouns, which are then used to learn extraction patterns. We also introduce a method to learn role-identifying expressions, which consist of a role-identifying verb linked to an event (e.g., “ participated in the murder”). We present experimental results on the MUC-4 terrorism corpus and a disease outbreaks corpus.