ParFDA for Instance Selection for Statistical Machine Translation

Ergun Biçici. ParFDA for Instance Selection for Statistical Machine Translation. In Proc. of the First Conference on Statistical Machine Translation (WMT16), Berlin, Germany, August 2016. Association for Computational Linguistics. [WWW] Keyword(s): Machine Translation, Machine Learning, Language Modeling.

We build parallel feature decay algorithms (ParFDA) Moses statistical machine translation (SMT) systems for all language pairs in the translation task at the first conference on statistical machine translation~\cite{WMT2016} (WMT16). ParFDA obtains results close to the top constrained phrase-based SMT with an average of 2.52 BLEU points difference using significantly less computation for building SMT systems than the computation that would be spent using all available corpora. We obtain BLEU bounds based on target coverage and show that ParFDA results can be improved by 12.6 BLEU points on average. Similar bounds show that top constrained SMT results at WMT16 can be improved by 8 BLEU points on average while German to English and Romanian to English translations results are already close to the bounds.