A study on improving turnover intention forecasting by solving imbalanced data problems: focusing on SMOTE and generative adversarial networks

Journal of Big Data

Table 3 Prediction analysis result

Set	Classifier	Accuracy			Precision			Recall			F1-score
Set	Classifier	R	S	G	R	S	G	R	S	G	R	S	G
Set1	LR	0.788	0.787	0.803	0.806	0.805	0.829	0.951	0.952	0.954	0.873	0.872	0.887
	KNN	0.769	0.771	0.800	0.808	0.810	0.826	0.915	0.914	0.938	0.858	0.859	0.878
	XGB	0.785	0.787	0.816	0.811	0.813	0.834	0.936	0.936	0.967	0.869	0.870	0.896
Set2	LR	0.770	0.770	0.799	0.782	0.782	0.828	0.958	0.957	0.944	0.861	0.861	0.882
	KNN	0.748	0.750	0.798	0.779	0.783	0.820	0.921	0.916	0.937	0.844	0.844	0.875
	XGB	0.772	0.774	0.829	0.789	0.790	0.843	0.945	0.947	0.954	0.860	0.861	0.895
Set3	LR	0.800	0.799	0.798	0.817	0.816	0.822	0.957	0.956	0.948	0.881	0.880	0.881
	KNN	0.778	0.779	0.794	0.819	0.823	0.818	0.917	0.912	0.932	0.865	0.865	0.871
	XGB	0.802	0.797	0.824	0.828	0.824	0.841	0.940	0.938	0.968	0.880	0.877	0.900
Set4	LR	0.776	0.774	0.796	0.787	0.785	0.818	0.958	0.958	0.941	0.864	0.863	0.875
	KNN	0.752	0.760	0.786	0.789	0.794	0.818	0.916	0.916	0.921	0.848	0.851	0.866
	XGB	0.781	0.781	0.824	0.796	0.797	0.836	0.946	0.946	0.955	0.865	0.865	0.892
Average	LR	0.783	0.782	0.799	0.798	0.797	0.824	0.956	0.955	0.937	0.870	0.869	0.877
	KNN	0.761	0.765	0.794	0.798	0.802	0.820	0.915	0.914	0.929	0.853	0.854	0.871
	XGB	0.785	0.784	0.823	0.806	0.806	0.838	0.942	0.941	0.951	0.869	0.868	0.891