The automatic extraction of bilingual single-word terms (SWTs) has been very successful, but for multi-word terms (MWTs), the precision is far from enough. This paper proposes a new approach for the automatic extraction of bilingual MWTs from a bilingual parallel corpus. We combine existing monolingual term extractor and a word alignment tool to extract bilingual terms. We introduced a re-segmentation to process a MWT as a single lexical unit so that it can be treated as a single unit by word alignment. We also improved the extraction rules of MWTs for existing term extractor and the experiment shows that our improvement is valid. We obtained a good precision and an improved BLEU score in our experiment based on a Chinese-Japanese parallel corpus.
展开▼
