Paper: An Exploration of Embeddings for Generalized Phrases

ACL ID P14-3006
Title An Exploration of Embeddings for Generalized Phrases
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014

Deep learning embeddings have been suc- cessfully used for many natural language processing problems. Embeddings are mostly computed for word forms although lots of recent papers have extended this to other linguistic units like morphemes and word sequences. In this paper, we define the concept of generalized phrase that in- cludes conventional linguistic phrases as well as skip-bigrams. We compute em- beddings for generalized phrases and show in experimental evaluations on corefer- ence resolution and paraphrase identifica- tion that such embeddings perform better than word form embeddings. 1 Motivation One advantage of recent work in deep learning on natural language processing (NLP) is that linguis- tic units are represented by rich and informative embeddings. These embeddings support b...