Table 3 Hyper-parameters in the JCLB
From: Joint contrastive learning and belief rule base for named entity recognition in cybersecurity
Parameters | Value |
|---|---|
Initial learning rate | 0.5 |
Embedded layer dimension | 768 |
LSTM dimension | 300 |
Epoch | 100 |
Batch size | 128 |
Dropout rate | 0.5 |
Learning rate | 0.001 |
Initial rule weight | 1 |
Initial premise weight | 0.5 |
Initial confidence | 0.1 |