NICT's Corpus Filtering Systems for the WMT18 Parallel Corpus Filtering Task
Abstract
This paper presents the NICT's participation in the WMT18 shared parallel corpus filtering task. The organizers provided 1 billion words German-English corpus crawled from the web as part of the Paracrawl project. This corpus is too noisy to build an acceptable neural machine translation (NMT) system. Using the clean data of the WMT18 shared news translation task, we designed several features and trained a classifier to score each sentence pairs in the noisy data. Finally, we sampled 100 million and 10 million words and built corresponding NMT systems. Empirical results show that our NMT systems trained on sampled data achieve promising performance.
Keywords
Cite
@article{arxiv.1809.07043,
title = {NICT's Corpus Filtering Systems for the WMT18 Parallel Corpus Filtering Task},
author = {Rui Wang and Benjamin Marie and Masao Utiyama and Eiichiro Sumita},
journal= {arXiv preprint arXiv:1809.07043},
year = {2018}
}
Comments
Due to the policy of our institute, with the agreement of all of the author, we decide to withdraw this paper