Table 1.
Comparison of performance evaluation of linear SVM and k-NN classifiers (using the Euclidian distance) for metabolic receptors classification in subcutaneous adipose tissue. The labeled examples include 52 positive examples and 55 negative examples (rows 1 and 2) and 61 cytokines receptors as negative examples (row 3).
| Method | Negative group | TP | TN | FP | FN | Sensitivity | Specificity | Accuracy | MCC | FP receptors | FN receptors | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | k-NN | Inferred receptors | 50 | 55 | 0 | 2 | 0.96 | 0.96 | 0.98 | 0.96 | NA | TFRC |
| TFR2 | ||||||||||||
| 2 | SVM | Inferred receptors | 44 | 55 | 0 | 8 | 0.85 | 0.87 | 0.93 | 0.86 | NA | ADIPOR2 |
| DRD4 | ||||||||||||
| EGFR | ||||||||||||
| FGFR2 | ||||||||||||
| LDLR | ||||||||||||
| LEPR | ||||||||||||
| TFR2 | ||||||||||||
| TFRC | ||||||||||||
| 3 | SVM/k-NN | Cytokines receptors | 42 | 60 | 1 | 10 | 0.81 | 0.86 | 0.9 | 0.81 | TNFRSF21 | ADIPOR2 |
| ADRA2B | ||||||||||||
| DRD4 | ||||||||||||
| EGFR | ||||||||||||
| FGFR2 | ||||||||||||
| FGFR4 | ||||||||||||
| LDLR | ||||||||||||
| LEPR | ||||||||||||
| TFR2 | ||||||||||||
| TFRC |