Table 3 Performance metrics of machine learning algorithms on balanced and imbalanced datasets using blind test evaluation.
From: A predictive model for damp risk in english housing with explainable AI
ML Algorithms | Data sets | Blind test performance | ||||
|---|---|---|---|---|---|---|
Accuracy | Precision | Recall (Sensitivity) | F-measure | AUC | ||
1. Neural Network | Balanced Data | 0.613 | 0.617 | 0.659 | 0.637 | 0.666 |
Imbalanced Data | 0.787 | 0.50 | 0.234 | 0.319 | 0.585 | |
2.Decision Tree | Balanced Data | 0.578 | 0.577 | 0.674 | 0.622 | 0.432 |
Imbalanced Data | 0.787 | 0 | 0 | 0 | 0.50 | |
3.XGBoost | Balanced Data | 0.585 | 0.592 | 0.628 | 0.610 | 0.604 |
Imbalanced Data | 0.787 | 0.50 | 0.280 | 0.359 | 0.701 | |
4.Random Forest | Balanced Data | 0.636 | 0.640 | 0.742 | 0.657 | 0.676 |
Imbalanced Data | 0.793 | 0.533 | 0.242 | 0.333 | 0.716 | |
5.Support Vector Machine (SVM) | Balanced Data | 0.656 | 0.619 | 0.863 | 0.721 | 0.656 |
Imbalanced Data | 0.784 | 0.40 | 0.030 | 0.056 | 0.645 | |
6.Logistic Regression | Balanced Data | 0.601 | 0.598 | 0.689 | 0.640 | 0.649 |
Imbalanced Data | 0.782 | 0.20 | 0.007 | 0.014 | 0.499 | |
7.Nearst Neighbours (KNN) | Balanced Data | 0.560 | 0.591 | 0.666 | 0.626 | 0.653 |
Imbalanced Data | 0.792 | 0.529 | 0.204 | 0.295 | 0.6 | |
