Table 8 The results of the second experiment in terms of AAC for MLMs, using the ICA algorithm in different cases of pre-processing combination
From: Prediction of diabetes disease using an ensemble of machine learning multi-classifier models
Preprocessing | Algorithm | N | k-NN | SVM | DT | RF | GNB | AdaBoost | Optimal MLM |
|---|---|---|---|---|---|---|---|---|---|
I + N | ICA | 4 | \(0.826\pm 0.011\) | \(0.890\pm 0.010\) | \(0.950\pm 0.004\) | \(0.951\pm 0.010\) | \(0.951\pm 0.003\) | \(0.930\pm 0.002\) | RF |
ICA | \(5\) | \(0.867\pm 0.012\) | \(0.900\pm 0.010\) | \(0.956\pm 0.005\) | \(0.958\pm 0.010\) | \(0.957\pm 0.002\) | \(0.931\pm 0.003\) | GNB | |
I + Z | ICA | 4 | \(0.880\pm 0.014\) | \(0.908\pm 0.004\) | \(0.959\pm 0.002\) | \(0.956\pm 0.005\) | \(0.962\pm 0.001\) | \(0.938\pm 0.004\) | GNB |
ICA | 5 | \(0.892\pm 0.011\) | \(0.910\pm 0.004\) | \(0.961\pm 0.006\) | \(0.961\pm 0.003\) | \(0.966\pm 0.003\) | \(0.941\pm 0.003\) | GNB |