Table 3 Accuracy and precision for proposed schemes using the vehicle dataset.
From: Algorithmic and mathematical modeling for synthetically controlled overlapping
Approach | Accuracy | Precision | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Method | 0% | 10% | 20% | 30% | 40% | 50% | 0% | 10% | 20% | 30% | 40% | 50% |
Majority class overlapping scheme (MOS) | ||||||||||||
 KNN | 75.29 | 74.71 | 66.29 | 65.57 | 62.57 | 53.97 | 75.12 | 74.44 | 65.39 | 64.79 | 62.69 | 55.00 |
 SVM | 76.47 | 82.76 | 75.84 | 68.31 | 67.91 | 69.63 | 77.05 | 82.46 | 75.25 | 71.19 | 69.80 | 67.57 |
 RF | 77.65 | 73.56 | 69.10 | 65.03 | 62.43 | 61.40 | 77.40 | 73.32 | 68.00 | 63.61 | 59.24 | 54.16 |
All class overlapping scheme (AOS) | ||||||||||||
 KNN | 75.29 | 63.10 | 55.67 | 51.36 | 45.99 | 43.31 | 75.12 | 63.13 | 57.29 | 51.50 | 45.75 | 43.20 |
 SVM | 76.47 | 66.84 | 72.41 | 62.73 | 56.12 | 57.87 | 77.05 | 66.95 | 72.47 | 61.88 | 54.22 | 58.64 |
 RF | 75.29 | 64.17 | 59.49 | 58.55 | 54.01 | 50.54 | 75.20 | 64.22 | 67.29 | 59.34 | 53.29 | 53.87 |
Random class overlapping scheme (ROS) | ||||||||||||
 KNN | 75.29 | 77.59 | 74.72 | 63.74 | 62.03 | 68.59 | 75.12 | 77.81 | 73.48 | 63.52 | 62.13 | 66.34 |
 SVM | 76.47 | 77.59 | 78.09 | 73.63 | 67.91 | 73.82 | 77.05 | 80.03 | 77.00 | 73.53 | 66.96 | 80.48 |
 RF | 77.06 | 78.74 | 77.53 | 74.73 | 61.50 | 70.16 | 77.40 | 79.70 | 76.93 | 74.77 | 60.07 | 68.61 |
SMOTE and all class overlapping scheme (AOS-SMOTE) | ||||||||||||
 KNN | 74.29 | 75.98 | 68.48 | 67.02 | 77.60 | 58.88 | 73.50 | 76.99 | 69.98 | 67.99 | 77.59 | 56.76 |
 SVM | 77.71 | 74.86 | 72.83 | 72.87 | 80.73 | 60.41 | 76.95 | 75.63 | 74.94 | 72.49 | 79.98 | 58.01 |
 RF | 79.43 | 75.42 | 75.00 | 69.68 | 80.21 | 59.90 | 79.40 | 76.04 | 77.61 | 69.55 | 79.65 | 57.80 |
