Table 5 Model parameter settings
From: Data and knowledge-driven named entity recognition for cyber security
Parameters | Setting |
|---|---|
Word embedding size | 100 |
Feature embedding size | 25 |
Word BiLSTM hidden node | 100 |
Feature BiLSTM hidden node | 25 |
Batch size | 32 |
Dropout rate | 0.5 |
Optimizier | Adam |
Learning rate | 0.001 |
Epochs | 15 |