Language Independent Connectivity Strength Features for Pivot Translation
Ahmed El Kholy and Nizar Habash
The 51st Annual Meeting of the Association for Computational Linguistics - Short Papers (ACL Short Papers 2013)
Sofia, Bulgaria, August 4-9, 2013
In this paper, we present language independent features to improve the quality of pivot-based statistical machine translation (SMT). We call this features connectivity strength features and they depend on the quality of projected alignments between the source and target phrases in the pivot phrase table. We show positive results on Persian-Arabic SMT as a case study.
Conference Manager (V2.61.0 - Rev. 2792M)