Research Article
All-Cause Death Prediction Method for CHD Based on Graph Convolutional Networks
Table 4
Coronary heart disease dataset prediction results.
| Index | Model | Accuracy (%) | AUC (%) | Precision (%) | Recall (%) | F1-score (%) |
| 1 | AdaBoost | 96.0 | 79.6 | 78.3 | 74.6 | 76.3 | 2 | Bayes | 94.3 | 84.5 | 67.2 | 65.3 | 66.2 | 3 | DT | 95.8 | 74.7 | 77.2 | 73.6 | 75.2 | 4 | SGD | 94.9 | 82.2 | 70.9 | 66.6 | 68.5 | 5 | SVC | 90.1 | 85.3 | 61.8 | 75.2 | 65.3 | 6 | DNN | 95.4 | 78.7 | 74.2 | 70.5 | 72.2 | 7 | p-GCN | 96.6 | 87.7 | 86.1 | 70.2 | 75.7 | 8 | K-GCN | 97.2 | 88.6 | 96.5 | 71.5 | 79.0 | 9 | AM-GCN | 97.3 | 90.4 | 93.4 | 74.4 | 80.9 |
|
|