Paper: Modeling and Learning Semantic Co-Compositionality through Prototype Projections and Neural Networks

ACL ID D13-1014
Title Modeling and Learning Semantic Co-Compositionality through Prototype Projections and Neural Networks
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2013
Authors

We present a novel vector space model for se- mantic co-compositionality. Inspired by Gen- erative Lexicon Theory (Pustejovsky, 1995), our goal is a compositional model where both predicate and argument are allowed to modify each others? meaning representations while generating the overall semantics. This readily addresses some major challenges with current vector space models, notably the pol- ysemy issue and the use of one represen- tation per word type. We implement co- compositionality using prototype projections on predicates/arguments and show that this is effective in adapting their word represen- tations. We further cast the model as a neural network and propose an unsupervised algorithm to jointly train word representations with co-compositionality. The model achieves the best res...