This paper describes a Verb Phrase Ellipsis (VPE) detection system, built for robustness, accuracy and domain independence. The system is corpus-based, and uses a variety of machine learning techniques on free text that has been automatically parsed using two different parsers. Tested on a mixed corpus comprising a range of genres, the system achieves a 72% F1-score. It is designed as the first stage of a complete VPE resolution system that is input free text, detects VPEs, and proceeds to find the antecedents and resolve them.
No takes yet. Share an insight, caveat, or question.
Leif Arda Nielsen (2004) studied this question.