The Automatic Interpretation of Nominalizations

This paper discusses the interpretation of nominalizations in domain independent wide-coverage text. We present a statistical model which interprets nominalizations based on the cooccurrence of verb-argument tuples in a large balanced corpus. We propose an algorithm which treats the interpretation task as a disambiguation problem and achieves a performance of approximately 80% by combining partial parsing, smoothing techniques and domain independent taxonomic information (e.g., WordNet).

• Type: Conference
• Year: 2000
• Where: AAAI
• Authors: Maria Lapata
• Keywords: AAAI 2000, Independent Taxonomic Information, Independent Wide-coverage Text, Intelligent Agents, Large Balanced Corpus