Table 2:

Detail descriptions about the performance measures.

Performance measures	Description
TP	When the positive samples are classified accurately
TN	When the negative samples are classified accurately
FP	When the negative samples are misclassified
FN	When the positive samples are misclassified
Accuracy	It is the overall classification accuracy percentage resulted by a standard classifier.
Sensitivity (Sen)	It determines the proportion of true positive samples in total samples and called as True Positive Rate (TPR) $\text{Sn}=TP/\left(TP+FN\right)$
Specificity (Spe)	It identifies the proportion of true negative samples in total samples and called as False Positive Rate (FPR) $\text{Sp}=TN/\left(TN+FP\right)$
Gmean	It is determined from both, specificity and sensitivity. Gmean = $\sqrt{\text{Sp}\times \text{Sn}}$
F-score	F-score value gives the combined performance of the two classes F-score = $\left(2\text{ × Sp × Sn}\right)/\left(\text{Sp }+\text{ Sn}\right)$
ROC	It is a graphical representation between sensitivity and specificity. ROC plots TPR versus FPR at different classification thresholds.
AUC	It is an aggregate evaluation of performance across all possible classification thresholds. So, it is called as Area under ROC curve.