Table 6 Results of the second experiment in terms of AAC for MLMs, using the MRMR feature selection algorithm in different modes of preprocessing 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 | Best MLM |
|---|---|---|---|---|---|---|---|---|---|
I | MRMR | 4 | \(0.931\pm 0.001\) | \(0.920\pm 0.001\) | \(0.916\pm 0.001\) | \(0.981\pm 0.001\) | \(0.922\pm 0.001\) | \(0.940\pm 0.001\) | RF |
MRMR | 6 | \(0.941\pm 0.002\) | \(0.924\pm 0.002\) | \(0.963\pm 0.002\) | \(0.986\pm 0.00\) 2 | \(0.936\pm 0.002\) | \(0.950\pm 0.002\) | RF | |
MRMR | 8 | \(0.930\pm 0.002\) | \(0.928\pm 0.002\) | \(0.950\pm 0.002\) | \(0.986\pm 0.00\) 2 | \(0.940\pm 0.002\) | \(0.954\pm 0.002\) | RF | |
MRMR | 10 | \(0.914\pm 0.002\) | \(0.932\pm 0.002\) | \(0.962\pm 0.002\) | \(0.985\pm 0.00\) 2 | \(0.946\pm 0.002\) | \(0.961\pm 0.002\) | RF | |
I + N | MRMR | 4 | \(0.969\pm 0.001\) | \(0.940\pm 0.002\) | \(0.944\pm 0.002\) | \(0.981\pm 0.00\) 2 | \(0.948\pm 0.002\) | \(0.940\pm 0.002\) | RF |
MRMR | 6 | \(0.958\pm 0.002\) | \(0.944\pm 0.002\) | \(0.935\pm 0.002\) | \(0.985\pm 0.00\) 2 | \(0.950\pm 0.002\) | \(0.952\pm 0.002\) | RF | |
MRMR | 8 | \(0.964\pm 0.003\) | \(0.931\pm 0.003\) | \(0.931\pm 0.003\) | \(0.986\pm 0.00\) 3 | \(0.952\pm 0.003\) | \(0.944\pm 0.003\) | RF | |
MRMR | 10 | \(0.971\pm 0.003\) | \(0.948\pm 0.003\) | \(0.958\pm 0.003\) | \(0.988\pm 0.00\) 3 | \(0.956\pm 0.003\) | \(0.946\pm 0.003\) | RF | |
I + Z | MRMR | 4 | \(0.931\pm 0.001\) | \(0.913\pm 0.001\) | \(0.940\pm 0.001\) | \(0.972\pm 0.00\) 1 | \(0.960\pm 0.001\) | \(0.942\pm 0.001\) | RF |
MRMR | 6 | \(0.930\pm 0.002\) | \(0.920\pm 0.002\) | \(0.947\pm 0.002\) | \(0.984\pm 0.00\) 2 | \(0.962\pm 0.001\) | \(0.940\pm 0.00\) 2 | RF | |
MRMR | 8 | \(0.930\pm 0.002\) | \(0.930\pm 0.002\) | \(0.948\pm 0.002\) | \(0.976\pm 0.00\) 2 | \(0.964\pm 0.003\) | \(0.943\pm 0.003\) | RF | |
MRMR | 10 | \(0.924\pm 0.003\) | \(0.916\pm 0.003\) | \(0.968\pm 0.003\) | \(0.977\pm 0.002\) | \(0.967\pm 0.003\) | \(0.945\pm 0.003\) | RF |