When people from different country contact with each other more and more nowadays, the need to understand or express a variety of information and knowledge in a language other than one's mother tongue grows dramatically. Accordingly, the research on Machine Translation, Multi-Language Information Retrieval and other related fields gets much attention, but these systems are still far from satisfaction. One of the most important reasons is the difficulty of acquiring translation knowledge. The traditional way to acquire translation knowledge by hand is very limited and costly. Therefore, extracting bilingual knowledge from a large number of real corpora has become a very important way and one of the key technologies is bilingual alignment.In this paper, first we implement a bilingual corpus search system based on keywords match on a clause-aligned Chinese-Japanese parallel corpus, which is a convenient tool for investigating the behavior of word alignment within a parallel sentence. Next, for word alignment, we do... |