Paper: Transducing Sentences to Syntactic Feature Vectors: an Alternative Way to Parse?

ACL ID W13-3205
Title Transducing Sentences to Syntactic Feature Vectors: an Alternative Way to Parse?
Venue Continuous Vector Space Models and their Compositionality
Session
Year 2013
Authors

Classification and learning algorithms use syntactic structures as proxies between source sentences and feature vectors. In this paper, we explore an alternative path to use syntax in feature spaces: the Dis- tributed Representation ?Parsers? (DRP). The core of the idea is straightforward: DRPs directly obtain syntactic feature vec- tors from sentences without explicitly pro- ducing symbolic syntactic interpretations. Results show that DRPs produce feature spaces significantly better than those ob- tained by existing methods in the same conditions and competitive with those ob- tained by existing methods with lexical in- formation.