Table 6.

The experimental performance of the proposed model (DNE-RL) vs other competitive deep learning benchmarks for Mendely Dataset.

Metric		GA	DE	PSO	GSK	MobileNet	VGGNet19	ResNet50	DenseNet12	DNE-RL
	AVG	0.927389	0.929467	0.936799	0.952668	0.956657	0.946883	0.961993	0.970338	0.987742
	STD	0.030495	0.029899	0.030885	0.028858	0.026337	0.026199	0.018996	0.017557	0.012116
ACC	Best	0.933583	0.948779	0.954091	0.962122	0.966641	0.965887	0.969985	0.978594	0.995186
	Worst	0.906488	0.909957	0.908777	0.934014	0.933933	0.930258	0.951332	0.960241	0.982168
	AVG	0.942783	0.938861	0.947751	0.948183	0.963662	0.956788	0.968848	0.973861	0.984334
	STD	0.025571	0.024167	0.023885	0.023955	0.021175	0.035537	0.023191	0.022818	0.013883
Precision	Best	0.956733	0.957883	0.959919	0.959192	0.975455	0.966881	0.976766	0.978588	0.991443
	Worst	0.916855	0.917919	0.920559	0.926634	0.939986	0.922065	0.941445	0.947766	0.984452
	AVG	0.921456	0.918872	0.925531	0.959983	0.948711	0.938636	0.956674	0.962678	0.989123
	STD	0.009886	0.009817	0.013553	0.011762	0.012835	0.011028	0.014224	0.016887	0.009931
Recall	Best	0.931441	0.923221	0.936637	0.963321	0.957766	0.945709	0.966838	0.968871	0.992172
	Worst	0.918965	0.914453	0.921298	0.949663	0.931655	0.925667	0.944913	0.948815	0.985561
	AVG	0.924881	0.927571	0.932889	0.948775	0.954686	0.943008	0.957814	0.968649	0.984939
	STD	0.002882	0.003662	0.003119	0.005891	0.018893	0.015944	0.019962	0.004881	0.002674
F-measure	Best	0.927644	0.931089	0.938875	0.095219	0.961008	0.949889	0.961129	0.971291	0.987175
	Worst	0.920307	0.922566	0.926618	0.944103	0.948871	0.939892	0.939072	0.965667	0.982323
	AVG	0.929331	0.928888	0.937676	0.951881	0.957117	0.945771	0.963442	0.969881	0.988466
	STD	0.017553	0.024554	0.015585	0.008914	0.005771	0.023319	0.019288	0.024596	0.015884
AUC	Best	0.932441	0.937669	0.942448	0.955676	0.960083	0.957119	0.972885	0.982011	0.991777
	Worst	0.925669	0.917765	0.932191	0.048911	0.953382	0.928228	0.948558	0.939006	0.981407