Table 10 Performance assessment of using the support vector machines (SVM) models.

From: Machine learning-based prediction of drug response in ischemia reperfusion animal model

Model

Support vector machines (SVM)

Metric

Complete

(all features)

Reduced model

(SFS)

Molecular

Accuracy

0.9047 ∓ 0.0340

0.6965 ∓ 0.0374

 

Precision

0.9907 ∓ 0.0131

0.8286 ∓ 0.0563

 

Recall

0.8700 ∓ 0.0384

0.7046 ∓ 0.0092

 

Specificity

0.9815 ∓ 0.0262

0.6833 ∓ 0.0970

 

MCC

0.8079 ∓ 0.0690

0.3636 ∓ 0.0926

Biochemical

Accuracy

0.8540 ∓ 0.0202

0.8932 ∓ 0.0084

 

Precision

0.9339 ∓ 0.0588

0.9508 ∓ 0.0398

 

Recall

0.8540 ∓ 0.0495

0.8942 ∓ 0.0493

 

Specificity

0.8630 ∓ 0.1238

0.8963 ∓ 0.0818

 

MCC

0.6924 ∓ 0.0568

0.7713 ∓ 0.0012

Molecular-biochemical

Accuracy

0.8708 ∓ 0.0210

0.8932 ∓ 0.0084

 

Precision

0.9431 ∓ 0.0618

0.9508 ∓ 0.0398

 

Recall

0.8700 ∓ 0.0384

0.8942 ∓ 0.0493

 

Specificity

0.8815 ∓ 0.1303

0.8963 ∓ 0.0818

 

MCC

0.7263 ∓ 0.0608

0.7713 ∓ 0.0012

  1. Results are reported in the (average accuracy ∓ variance) format for all the 3-folds used to cross-validate the models. Note that recall is also the sensitivity metric. MCC, Matthews correlation coefficient.
Back to article page