Selection Restrictions Acquisition for Parsing and Information Retrieval Improvement.
Alexandre Agustini, Pablo Gamallo, Gabriel Pereira Lopes · 2001
Natural language parsing requires extensive lexicons containing subcategorization information for speci c sublanguages. This paper describes an automatic clustering strategy for acquiring selection restrictions from corpora. In order to test our method, preliminary experiments have been performed on a law-case domain Portuguese corpus. The acquired information is used to lexicon tuning for parsing improvement.