Table 2.
Comparison of the model developed in this work with the existing classifiers. FN: Number of features used to build the classification model, ML: machine learning method, SE: sensitivity, SP: specificity, Acc: accuracy, SVM: support vector machine, NB: naïve Bayes, MLP: multilayered perceptron, RF: random forest, DT: decision tree.
| Method | FN | ML | SE | SP | Acc |
|---|---|---|---|---|---|
| Triplet-SVM (Xue et al., 2005) | 32 | SVM | 93.30 | ||
| MiPred (Jiang et al., 2007) | 34 | RF, SVM | 98.21 | 95.09 | 96.68 |
| MicroPred (Batuwita et al., 2009) | 21 | RF, SVM | 90.02 | 97.28 | |
| izMiR (Saçar Demirci et al., 2017) | ~900 | SVM, NB, DT | 91.98 | 91.98 | 91.25 |
| 3D model | 36 | RF, NB, DT | 98.87 | 98.87 | 98.58 |