Classification of lung cancer severity using gene expression data based on deep learning

Table 7 Accuracy evaluation of classical ML models using different training-testing methods with 3000 features

Dataset	Training – testing method	SVM	RF	KNN	AdaBoost
LUAD	10-fold cross-validation	0.71	0.58	0.58	0.58
	5-fold cross-validation	0.70	0.61	0.59	0.58
	80–20 splitting	0.79	0.63	0.61	0.67
	70–30 splitting	0.78	0.63	0.61	0.61
LUSC	10-fold cross-validation	0.711	0.63	0.60	0.58
	5-fold cross-validation	0.719	0.59	0.60	0.59
	80–20 splitting	0.76	0.67	0.67	0.62
	70–30 splitting	0.69	0.66	0.61	0.58

ISSN: 1472-6947