Automatic induction of FrameNet lexical units

Most attempts to integrate FrameNet in NLP systems have so far failed because of its limited coverage. In this paper, we investigate the applicability of distributional and WordNetbased models on the task of lexical unit induction, i.e. the expansion of FrameNet with new lexical units. Experimental results show that our distributional and WordNet-based models achieve good level of accuracy and coverage, especially when combined.

• Type: Conference
• Year: 2008
• Where: EMNLP
• Authors: Marco Pennacchiotti, Diego De Cao, Roberto Basili, Danilo Croce, Michael Roth
• Keywords: EMNLP 2008, Lexical Unit Induction, Lexical Units, Most Attempts, Natural Language Processing