next up previous
次へ: Decoder 上へ: Experiments with our Machine 戻る: -gram Language Model

Development Data

We used 8096 sentences for development data (called IWSLT10.devset3_IIWSLT05.mref) for French-English SMT. Out of these 8096 sentences, 1375 sentences matched with the French-English patterns. For the results, we obtained 1375 ``English'' sentences. These 1375 ``English'' sentences were used as development data for ``English''-English translation.



Jin'ichi Murakami 平成22年12月20日