Optimizing prognostic factors of five-year survival in gastric cancer patients using feature selection techniques with machine learning algorithms: a comparative study

Table 8 Time to build the model, correctly classified instances, incorrectly classified instances, and accuracy of the ML models

Feature selection algorithm used	NONE			Boruta-F			mRMR-F			LASSO-F			Relief –F
Classifier	XGB	HGB	SVM	XGB	HGB	SVM	XGB	HGB	SVM	XGB	HGB	SVM	XGB	HGB	SVM
Time to build a model (s)	3081	3301	2849	2806	2930	2200	2624	2580	2147	2490	2028	1981	2924	1914	1280
Correctly classified instance	67	60	65	83	85	80	79	79	77	80	81	86	80	79	81
Incorrectly classified instance	30	37	32	14	12	17	18	18	20	17	16	11	17	18	16
Accuracy	69.47	62.58	68.25	85.68	88.25	82.54	82.12	81.46	80.24	83.07	84.12	89.10	83.82	82.47	83.75

ISSN: 1472-6947