RLMD-PA: A Reinforcement Learning-Based Myocarditis Diagnosis Combined with a Population-Based Algorithm for Pretraining Weights

Table 9

Performance evaluation obtained for various values of as the reward of the majority class.

	Accuracy	Recall	Precision	F-measure	Specificity	G-means
0	0.807	0.778	0.727	0.752	0.824	0.801
0.1	0.838	0.814	0.769	0.791	0.853	0.833
0.2	0.867	0.844	0.810	0.827	0.880	0.862
0.3	0.884	0.858	0.837	0.847	0.900	0.879
0.4	0.877	0.848	0.830	0.839	0.895	0.871
0.5	0.857	0.814	0.807	0.810	0.883	0.848
0.6	0.845	0.798	0.792	0.795	0.874	0.835
0.7	0.825	0.764	0.768	0.766	0.861	0.811
0.8	0.807	0.738	0.746	0.742	0.848	0.791
0.9	0.792	0.709	0.730	0.719	0.842	0.773
1	0.779	0.695	0.710	0.702	0.829	0.759