Table 8.

Evaluation results of AMD-Neov dataset using different sampling strategies with three metrics across ten classification algorithms.

Method	Metric	Classifier
		J48	3NN	NB	RF5	LOG	1NN	7NN	SMO	RF10	RBFNet	R. Avg.
PSO	AUC	0.681	0.659	0.661	0.662	0.678	0.656	0.694	0.628	0.686	0.672	0.668
	FMeasure	0.549	0.557	0.537	0.566	0.545	0.556	0.572	0.559	0.552	0.559	0.555
	GMean	0.622	0.628	0.619	0.643	0.626	0.630	0.648	0.637	0.631	0.631	0.632
	C. Avg.	0.617	0.615	0.605	0.624	0.616	0.614	0.638	0.608	0.623	0.621	0.618
RU	AUC	0.652	0.627	0.625	0.622	0.635	0.649	0.622	0.619	0.663	0.631	0.635
	FMeasure	0.549	0.526	0.524	0.534	0.519	0.531	0.543	0.529	0.561	0.539	0.536
	GMean	0.637	0.602	0.601	0.609	0.596	0.615	0.615	0.604	0.636	0.612	0.613
	C. Avg.	0.613	0.585	0.583	0.588	0.583	0.598	0.593	0.584	0.620	0.594	0.595
RO	AUC	0.643	0.643	0.646	0.659	0.635	0.655	0.638	0.632	0.660	0.657	0.647
	FMeasure	0.507	0.542	0.491	0.516	0.498	0.516	0.521	0.506	0.534	0.531	0.516
	GMean	0.602	0.629	0.589	0.610	0.599	0.612	0.612	0.598	0.624	0.623	0.610
	C. Avg.	0.584	0.605	0.575	0.595	0.577	0.594	0.590	0.579	0.606	0.603	0.591
Cluster	AUC	0.656	0.624	0.627	0.629	0.625	0.652	0.644	0.594	0.642	0.638	0.633
	FMeasure	0.551	0.524	0.502	0.538	0.506	0.521	0.546	0.504	0.536	0.537	0.527
	GMean	0.641	0.605	0.587	0.624	0.591	0.610	0.630	0.585	0.620	0.621	0.611
	C. Avg.	0.616	0.584	0.572	0.597	0.574	0.594	0.607	0.561	0.599	0.599	0.590