Table 9 Comparative performance of ensemble machine learning models on various micro-averaging measures
From: Prediction of diabetes disease using an ensemble of machine learning multi-classifier models
EMLMs | Precision | Recall | F1-Score | Accuracy | AUC |
|---|---|---|---|---|---|
I + N + K-NN + RF | \(0.910\pm 0.002\) | \(0.857\pm 0.002\) | \(0.877\pm 0.002\) | \(0.996\pm 0.0001\) | \(0.974\pm 0.002\) |
I + K-NN + GNB + RF | \(0.903\pm 0.002\) | \(0.870\pm 0.00\) 5 | \(0.870\pm 0.410\) | \(0.998\pm 0.0000\) | \(0.965\pm 0.006\) |
I + K-NN + AB + DT + RF | \(0.986\pm 0.001\) | \(0.979\pm 0.002\) | \(0.985\pm 0.001\) | \(0.998\pm 0.0007\) | \(0.999\pm 0.000\) |
I + K-NN + GNB + RF + DT + AB | \(0.940\pm 0.002\) | \(0.873\pm 0.006\) | \(0.897\pm 0.010\) | \(0.998\pm 0.0003\) | \(0.988\pm 0.001\) |
I + K-NN + GNB + RF + DT + AB + SVM | \(0.940\pm 0.002\) | \(0.910\pm 0.001\) | \(0.903\pm 0.160\) | \(0.998\pm 0.0003\) | \(0.980\pm 0.004\) |