Table 16 Performance summary of models with and without feature selection.
From: Predicting land suitability for wheat and barley crops using machine learning techniques
Models | No features | Accuracy (%) | Precision (%) | Recall (%) | F1-score (%) |
|---|---|---|---|---|---|
RF | 16 | 98.76 | 98.57 | 98.7 | 98.61 |
RF with UFS | 13 | 97.43 | 96.96 | 97.14 | 97.0 |
RF with RFECV | 13 | 98.76 | 98.58 | 98.71 | 98.63 |
RF with SFS | 9 | 98.96 | 98.93 | 98.89 | 98.88 |
GB | 16 | 99.01 | 99.0 | 98.9 | 98.93 |
GB with UFS | 14 | 97.73 | 97.32 | 97.3 | 97.28 |
GB with RFECV | 6 | 99.06 | 99.02 | 98.93 | 98.96 |
GB with SFS | 10 | 99.41 | 99.37 | 99.34 | 99.35 |
KNN | 16 | 96.74 | 96.3 | 96.4 | 96.3 |
KNN with UFS | 14 | 96.64 | 96.14 | 96.25 | 96.14 |
KNN with SFS | 6 | 97.33 | 96.97 | 96.98 | 96.92 |
