Classifier Combination for Contextual Idiom Detection Without Labelled Data

We propose a novel unsupervised approach for distinguishing literal and non-literal use of idiomatic expressions. Our model combines an unsupervised and a supervised classifier. The former bases its decision on the cohesive structure of the context and labels training data for the latter, which can then take a larger feature space into account. We show that a combination of both classifiers leads to significant improvements over using the unsupervised classifier alone.

• Type: Journal
• Year: 2009
• Where: EMNLP
• Authors: Linlin Li, Caroline Sporleder
• Keywords: EMNLP 2009, Idiomatic Expressions, Larger Feature Space, Natural Language Processing, Unsupervised Classifier