Skip to main content

Table 4 the overall classification performance comparison of 9 classifiers averaged in macro and micro level

From: Semantic categorization of Chinese eligibility criteria in clinical trials using machine learning methods

Models Macro-average Micro-average   
Precision Recall F1-score Precision Recall F1-score
Machine learning algorithms       
 NB 0.5398 0.7403 0.5965 0.6312 0.6312 0.6312
 kNN 0.7531 0.6693 0.6948 0.7632 0.7632 0.7632
 LR 0.8017 0.7574 0.7732 0.8173 0.8173 0.8173
 SVM 0.8196 0.7712 0.7899 0.8293 0.8293 0.8293
Deep learning algorithms       
 CNN 0.8004 0.6951 0.7258 0.8142 0.8142 0.8142
 RNN 0.7837 0.6925 0.7170 0.8138 0.8138 0.8138
 FastText 0.7645 0.7188 0.7341 0.8182 0.8182 0.8182
Pre-trained language models       
 BERT 0.7994 0.8023 0.7958 0.8447 0.8447 0.8447
 ERNIE 0.7964 0.8074 0.7980 0.8484 0.8484 0.8484
  1. Bold indicates the best value per metric