Table 7 Results of test dataset for distant recurrence model
Particular | Data | ||
|---|---|---|---|
Model name | Distant recurrence | ||
No. of samples (rows) | 311 | 311 | 311 |
Total number of independent variables (columns) | 388 | 388 | 388 |
ML algorithm used | Random Forest | KSVM | XGBoost |
Feature selection technique | SFFS | SFFS | SFFS |
OverSampling method used | ADASYN | ADASYN | ADASYN |
No. of samples after OverSampling | 573 | 573 | 573 |
Number of synthetic samples | 262 | 262 | 262 |
No. of independent variables used in the dataset | 22 | 18 | 86 |
Mean accuracy train score | 0.99 | 0.99 | 1 |
Mean accuracy test score | 0.87 | 0.95 | 0.87 |
Sensitivity | 0.92 | 0.90 | 0.75 |
Specificity | 0.93 | 0.98 | 0.81 |
Mean F-Score train label 0 | 0.99 | 0.99 | 1 |
Mean F-Score train label 1 | 0.99 | 0.99 | 1 |
Mean F-score test label 0 | 0.92 | 0.96 | 0.92 |
Mean F-score test label 1 | 0.68 | 0.91 | 0.64 |
No. of samples for class-0 | 281 | 281 | 281 |
No. of samples for class-1 | 30 | 30 | 30 |
Base algorithm | Random forest | KSVM | XGBoost |
ROC_AUC_Score | 0.96 | 0.99 | 0.97 |