In this paper, we utilize segmentation alternatives. Our research contribution is a novel estimation method of the translation probabilities used in phrase-based statistical machine translation in order to reflect the trustworthiness of the segmentation. Our system, however, underperforms the baseline.
Cite as: Na, H., Lee, J.-H. (2010) The POSTECH's statistical machine translation system for the IWSLT 2010. Proc. International Workshop on Spoken Language Translation (IWSLT 2010), 153-156
@inproceedings{na10_iwslt,
author={Hwidong Na and Jong-Hyeok Lee},
title={{The POSTECH's statistical machine translation system for the IWSLT 2010}},
year=2010,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2010)},
pages={153--156}
}