Table 3 Model performance with ICA + PCA dimensionality reduction and bayesian fusion across balancing methods.
From: Enhanced cervical cancer diagnosis using a novel Bayesian fusion ensemble method with explainable AI
Exp. No | Method used | Model Name | Accuracy (%) | Precision | Recall | F1-value | AUC-ROC | CV = 5 | CV = 10 |
|---|---|---|---|---|---|---|---|---|---|
01 | SMOTE + ICA + PCA | RF | 94.19 | 0.95 | 0.94 | 0.94 | 0.95 | 95.39 | 95.22 |
DT | 92.44 | 0.95 | 0.92 | 0.93 | 0.83 | 93.34 | 93.15 | ||
NB | 87.21 | 0.94 | 0.87 | 0.90 | 0.93 | 88.65 | 88.12 | ||
ANN | 95.35 | 0.95 | 0.95 | 0.94 | 0.88 | 96.44 | 96.32 | ||
XGB | 94.19 | 0.95 | 0.94 | 0.94 | 0.94 | 96.57 | 96.61 | ||
KNN | 95.11 | 0.95 | 0.96 | 0.96 | 0.89 | 94.33 | 94.52 | ||
Proposed model | 95.93 | 0.71 | 0.69 | 0.96 | 0.97 | 96.95 | 96.88 | ||
02 | ROS + ICA + PCA | RF | 98.76 | 0.98 | 0.99 | 0.99 | 1.00 | 97.34 | 96.98 |
DT | 96.89 | 0.93 | 1.00 | 0.96 | 0.97 | 95.53 | 95.94 | ||
NB | 75.78 | 0.61 | 0.93 | 0.74 | 0.87 | 77.78 | 77.12 | ||
ANN | 96.68 | 0.93 | 1.00 | 0.96 | 0.99 | 94.44 | 94.32 | ||
XGB | 97.72 | 0.95 | 1.00 | 0.97 | 0.98 | 96.38 | 96.51 | ||
KNN | 95.85 | 0.92 | 1.00 | 0.94 | 0.98 | 92.01 | 92.20 | ||
Proposed model | 99.88 | 0.99 | 1.00 | 0.99 | 1.00 | 99.67 | 99.50 |
