A contrastive study of Chinese progressive aspect structure, "ZAI+verb" and its English correspondents: a bilingual parallel corpus-based perspective

One of the Chinese-English machine translation burning problems is how to convert the Chinese aspects to their English correspondents. Chinese progressive aspect marker "ZAI+verb" can have multiple English correspondents besides the English progressive aspect. This paper takes an empirical approach to this unresolved conversion, based on a bilingual parallel corpus. It identifies 10 English translation equivalents of "ZAI+verb", figures out the conversion rates between "ZAI+verb" and its English translation equivalents, and presents 14 conversion rules to define the matching relationships between them. The results implies that the precise degree of Chinese-English machine translation of "ZAI+verb" to progressive aspects in English can be expected to be up to 67%.