Table 2 Performance of utilizing variable numbers of clinical features in different classifiers to predict diabetic nephropathy.
Classifier | No. of features | Accuracy (%) | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|
Decision tree | 7 | 62.17 | 55.56 | 69.02 |
Random forest | 6 | 63.91 | 60.68 | 67.26 |
SVM | 3 | 60.87 | 50 | 78.72 |
Naïve Bayes | 7 | 62.61 | 41.03 | 84.96 |