次へ: Discussion
上へ: Human Evaluation Results
戻る: Simple Sentences
We selected 10 sentences at random for each rank. The results of the human evaluation for compound/complex sentences are listed in Table XVIII.
表 XVIII:
Results of Human Evaluation(compound/complex sentences)
|
|
Table XVIII indicates the proposed method is superior to the baseline(Moses) for the A-rank and the B-rank. On the other hand, its results were split for the C-rank, and it was inferior to the baseline(Moses) for the D-rank. This shows the effectiveness of the proposed method for the A-rank and the B-rank.
平成25年9月17日
![\scalebox{1.0}[1.0]{
\begin{tabular}{\vert l\vert c\vert c\vert c\vert c\vert}
\...
...e
C-rank & 2 & 2 & 5 & 1\\ \hline
D-rank & 0 & 2 & 8 & 0\\ \hline
\end{tabular}}](img28.png)