We used 10-fold cross validation to evaluate the classification models. The performance of classifiers is evaluated with overall accuracy (Acc), sensitivity (Sen), specificity (Spe), Kappa statistics (Ks), precision, and F-score, as shown in Equations (10)–(15). (10)Acc =(TP + TN)(TP + TN + FP + FN)×100% (11)Sen =TPTP + FN×100% (12)Spe =TNTN + FP×100% (13)Ks =Pa−Pb 1−Pb (14)F−Score=2×TP(2×TP+FP+FN)×100% (15)Precision =TPTP+FP ×100% where TP is a true positive, TN is a true negative, FP is a false positive, and FN is a false negative. Pa shows the observed proportion of agreement, and Pb indicates the proportion of agreement expected by chance in Equation (13).
