Table 15. Comparison of the proposed work with previous work.

Classifier	No. of features	Accuracy(%)	Recall(%)	F1-score(%)
kNN [16]	6	85.32	90.24	--
random forest [16]	6	79.57	87.50	--
Naïve Bayes [17]	--	77.79	77.7	78.40
random forest [17]	--	81.93	81.9	81.90
Naïve Bayes, MLP [17]	--	83.87	83.9	84.20
random forest, MLP [17]	--	80.00	80.00	80.10
Decision table, random forest [17]	--	81.29	81.30	81.30
Bayes Net, MLP [17]	--	81.93	81.90	82.10
random forest, Naïve Bayes [17]	--	78.38	78.40	79.00
SVM—linear kernel, training size 40% [3]	6	85.00	--	--
SVM -KMOD kernel, training size 40% [3]	6	83.90	--	--
SVM -linear kernel, training size 80% [3]	6	84.30	--	--
SVM -KMOD kernel, training size 80% [3]	6	85.90	--	--
Bagging SVM–linear kernel with c = 7, and holdout [Proposed work]	5	86.96	95.92	91.26
Bagging SVM–linear kernel with c = 7, and 28 fold cross validation [Proposed work]	5	86.94	91.43	90.14