TABLE 3.

Wrapper-based Supervised Gene Selection.

References	Ideology	Gene Selection Algorithm	Classifier	Dataset	Performance Evaluation Metrics
Wang A. et al., 2017	Aims to improve the evaluation time with the help of Markov Blanket with Sequential Forward Selection.	Wrapper-based Sequential Forward Selection with Markov Blanket	kNN, Naïve Bayes, C4.5 Decision Tree	• Colon • SRBCT • Leukemia • DLBCL • Prostate • Bladder • Gastric • Tox • Blastoma	• Classification Accuracy • Wilcoxon signed-rank test
Hasri et al., 2017	The proposed method Multiple Support Vector Machine – Recursive Feature Elimination is an enhancement of SVM-RFE for improving the accuracy in selecting the informative features.	MSVM-RFE	Random Forest, C4.5 Decision Tree	• Leukemia • Lung Cancer	• Classification Accuracy
Shanab et al., 2014	A wrapper-based feature selection technique has been developed with Naïve Bayes by using the real-world high dimensional data in terms of difficulty due to noise.	Naïve Bayes-Wrapper	Naïve Bayes, MLP, 5NN, SVM and Logistic Regression	• Ovarian • ALL AML Leukemia • CNS • Prostate MAT • Lymphoma • Lung Cancer	• AUC
Mishra and Mishra, 2015	This method aims to gather the relevant genes to distinguish the biological facts. The method is an extension of SVM-T-RFE, where instead of a t-test, a Bayesian t-test has been used for better results.	SVM- Bayesian T-Test –RFE (SVM-BT-RFE)	SVM-RFE, SVM-T-RFE	• Colon • Leukemia • Medulla Blastoma • Lymphoma • Prostate	• Classification Accuracy
Zhang et al., 2018	In this study, three wrapper based feature selections are implemented, and the results show that SVM-RFE-PSO performs better in selecting informative features than the other two.	SVM-RFE-GS, SVM-RFE-PSO, and SVM-RFE-GA	SVM	• Breast Cancer • TGCA	• AUC • Accuracy • Precision • Recall • F-Score