Table 4 The performance comparison of different classifiers in the Classifier Chain and Binary Relevance methods
Method | Model | Accuracy | Macro Recall | Macro Precision | Macro F1-Score | Macro AUC |
|---|---|---|---|---|---|---|
Binary Relevance | Decision Tree | 0.63 | 0.70 | 0.71 | 0.70 | 0.80 |
Classifier Chain | Decision Tree | 0.71 | 0.70 | 0.71 | 0.70 | 0.80 |
Binary Relevance | KNN | 0.72 | 0.69 | 0.83 | 0.74 | 0.90 |
Classifier Chain | KNN | 0.73 | 0.69 | 0.82 | 0.74 | 0.90 |
Binary Relevance | Logistic Regression | 0.26 | 0.31 | 0.55 | 0.39 | 0.74 |
Classifier Chain | Logistic Regression | 0.45 | 0.49 | 0.54 | 0.48 | 0.71 |
Binary Relevance | Neural Network | 0.46 | 0.42 | 0.62 | 0.47 | 0.80 |
Classifier Chain | Neural Network | 0.57 | 0.49 | 0.62 | 0.51 | 0.80 |
Binary Relevance | Random Forest | 0.73 | 0.73 | 0.80 | 0.76 | 0.93 |
Classifier Chain | Random Forest | 0.76 | 0.75 | 0.77 | 0.76 | 0.90 |
Binary Relevance | SVM | 0.04 | 0.02 | 0.17 | 0.04 | 0.70 |
Classifier Chain | SVM | 0.04 | 0.02 | 0.17 | 0.04 | 0.73 |
