TÜBİTAK SMT System Submission for WMT2016

By Emre Bektas, Ertugrul Yilmaz, Coskun Mermer, Ilknur Durgar El-Kahlout. 2016

3 citations recorded by the source corpus.

We describe the T ¨UB˙ITAK Turkish-English machine translation systems sub-missions in both directions for the WMT 2016: News Translation Task. We exper-iment with phrase-based and hierarchical phrase-based systems for both directions using word-level and morpheme-level representations for the Turkish side. Finally we perform system combination which re-sults in 0.5 BLEU increase for Turkish-to-English and 0.3 BLEU increase for English-to-Turkish.

Subfields: Artificial Intelligence