Research Article
An Approach Based on Multilevel Convolution for Sentence-Level Element Extraction of Legal Text
Table 9
The comparison results on legal dataset of the concatenated method of outputs of different CNN layers.
| | Model | Divorce | Loan | Labor | | HL (-) | Micro-P (+) | Micro-R (+) | Micro-F1 (+) | HL (-) | Micro-P (+) | Micro-R (+) | Micro-F1 (+) | HL (-) | Micro-P (+) | Micro-R (+) | Micro-F1 (+) |
| | w/o | 0.02257 | 0.8676 | 0.8591 | 0.8633 | 0.02910 | 0.8479 | 0.8385 | 0.8432 | 0.01876 | 0.8599 | 0.8301 | 0.8447 | | Our model | 0.02126 | 0.8852 | 0.8728 | 0.8790 | 0.02772 | 0.8633 | 0.8573 | 0.8603 | 0.01688 | 0.8812 | 0.8589 | 0.8699 |
|
|
