Simple Sentences

We input 10,000 English sentences and obtained 3,252 sentences matching the English-Japanese translation patterns. We obtained 379 sentences in the A-rank, 247 in the B-rank, 292 in the C-rank, and 2,334 in the D-rank. And we compared our method with the baseline(Moses) for each rank. The automatic evaluation results are listed in Table XV.

表 XV: Automatic Evaluation Results

From the results in Table XV, we can see the BLEU and NIST values were higher for the A-rank. This means the proposed method was more effective than the baseline(Moses) for the A-rank. However, the same cannot be said for the other ranks.