Adapting Decision Trees for Learning Selectional Restrictions

Sean Szumlanski, Fernando Gomez

This paper describes the implementation of a system that automatically learns selectional restrictions for individual senses of polysemous verbs from subject-object relationships. The selectional restrictions are inferred from an adaptation of decision tree induction, and are bound to the syntactic relations that realize them as part of a move toward automated construction of verb predicates.

Subjects: 13. Natural Language Processing; 12. Machine Learning and Discovery

Submitted: Feb 25, 2008

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.