Quantitative evaluation of English-Japanese machine translation of medical literature.

Although many machine-translation programs are currently available, few evaluation methods of such translation exist for any given application area. It is difficult to evaluate machine-translation systems objectively because the quality of a translation depends on the combination of three factors: the translation program, the dictionary, and the original document. In this study, we developed a quantitative evaluation method for assessing machine translation, which evaluates these three factors separately. We applied this method to the translation of English to Japanese for medical literature and the method proved to be a good indicator for further system improvement. Using this method we also discovered other important points for machine translation, such as the examination of target documents for the construction of a better application dictionary.