Research Article
Unsupervised Quality Estimation Model for English to German Translation and Its Application in Extensive Supervised Evaluation
Table 6
The performances on WMT12 testing corpora using Spearman correlation.
| System | Correlation score with human judgments | Other-to-English | English-to-other | Mean | CS-EN | DE-EN | ES-EN | FR-EN | Mean | EN-CS | EN-DE | EN-ES | EN-FR | Mean |
| | 0.89 | 0.77 | 0.91 | 0.81 | 0.85 | 0.75 | 0.34 | 0.45 | 0.77 | 0.58 | 0.71 | METEOR | 0.66 | 0.89 | 0.95 | 0.84 | 0.84 | 0.73 | 0.18 | 0.45 | 0.82 | 0.55 | 0.69 | BLEU | 0.89 | 0.67 | 0.87 | 0.81 | 0.81 | 0.80 | 0.22 | 0.40 | 0.71 | 0.53 | 0.67 | TER | 0.89 | 0.62 | 0.92 | 0.82 | 0.81 | 0.69 | 0.41 | 0.45 | 0.66 | 0.55 | 0.68 |
|
|
Bold fonts mean the best performance.
|