Research Article
Identification of Dry Bean Varieties Based on Multiple Attributes Using CatBoost Machine Learning Algorithm
Table 4
Performance metrics of proposed MLAs for the seven varieties of dry beans.
| ML classifiers | Classes | Precision | Recall | f1-score |
| Random forest | Sira | 0.85 | 0.83 | 0.84 | Bombay | 1 | 1 | 1 | Dermason | 0.92 | 0.91 | 0.91 | Barbunya | 0.93 | 0.94 | 0.94 | Horoz | 0.94 | 0.95 | 0.94 | CAli | 0.94 | 0.95 | 0.95 | Seker | 0.95 | 0.94 | 0.94 |
| XGBoost | Sira | 0.83 | 0.84 | 0.84 | Bombay | 1 | 1 | 1 | Dermason | 0.93 | 0.89 | 0.91 | Barbunya | 0.95 | 0.95 | 0.95 | Horoz | 0.93 | 0.96 | 0.94 | Cali | 0.95 | 0.97 | 0.96 | Seker | 0.95 | 0.93 | 0.94 |
| CatBoost | Sira | 0.88 | 0.84 | 0.86 | Bombay | 1 | 1 | 1 | Dermason | 0.94 | 0.94 | 0.94 | Barbunya | 0.94 | 0.94 | 0.94 | Horoz | 0.94 | 0.97 | 0.95 | Cali | 0.94 | 0.97 | 0.95 | Seker | 0.95 | 0.93 | 0.94 |
|
|