Table 3 Per value results under different vectorization methods.
From: Scoring method of English composition integrating deep learning in higher vocational colleges
Composition collection | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
BERT | 0.156 | 0.132 | 0.413 | 0.151 | 0.561 |
BERT-Sentence | 0.540 | 0.561 | 0.160 | 0.256 | 0.304 |
Fast-Text | 0.133 | 0.242 | 0.406 | 0.107 | 0.521 |
BERT-Flow | 0.553 | 0.572 | 0.233 | 0.285 | 0.342 |