Word Order Correction for Language Transfer Using Relative Position Language Modeling

Sentence correction has been an important and emerging issue in computer-assisted language learning. However, existing techniques based on grammar rules or statistical machine translation are still not robust enough to tackle the common incorrect word order errors in sentences produced by second language learners of Chinese. In this paper, a novel relative position language model is proposed to address this problem, for which a corpus of erroneous English-Chinese language transfer sentences along with their corrected counterparts is created and manually judged by human annotators. Experimental results show that compared to a scoring approach based on an n-gram language model and a phrase-based machine translation system, the performance in terms of BLEU scores of the proposed approach achieved improvements of 20.3% and 26.5% for the correction of word order errors resulting from language transfer, respectively.