Parsing Statistical Machine Translation Output
Simon Carter and Christof Monz
In: LTC 2009, August 1-3, 2009, Poland.

Abstract

Despite increasing research into the use of syntax during statistical machine translation, the incorporation of syntax into language models has seen limited success. We present a study of the discriminative abilities of generative syntax-based language models, over and above standard n-gram models, with a focus on potential applications for statistical machine translation. We show that a relatively simple parsing model based on Greibach Normal Form can outperform established state of the art parsers in discriminating between well-formed and un-grammatical English.

[Edit]