Table 5.

The classification accuracy was obtained with the XGBoost algorithm for each subject in the test set. The shaded green cells with bold text show the highest accuracy obtained for each subject. At the bottom, the best mean accuracy for each ratio index is marked in the same way.

Subject	Index1	Index2	θ/α	β/α	(θ + α)/β	θ/β	(θ + α)/(α + β)	γ/δ	(γ + β)/(δ + α)
	This Work		Eoh et al. [36]		Jap et al. [37]			da Silveira et al. [10]
tr04-0653	0.5672	0.6345	0.6303	0.5042	0.5168	0.5840	0.5630	0.5612	0.5738
tr05-0028	0.4454	0.4202	0.5294	0.5588	0.4664	0.4076	0.5462	0.5021	0.5443
tr05-0332	0.8067	0.8277	0.7563	0.5630	0.5294	0.6555	0.6134	0.5654	0.6118
tr07-0127	0.5462	0.6092	0.4916	0.5294	0.5252	0.5000	0.4118	0.4304	0.4051
tr08-0157	0.6303	0.6681	0.5294	0.5294	0.5000	0.5084	0.5294	0.5232	0.4810
tr09-0328	0.6050	0.5084	0.5966	0.5168	0.5000	0.5042	0.5672	0.5738	0.5401
tr09-0453	0.5588	0.5252	0.5714	0.5420	0.5294	0.5504	0.5840	0.5105	0.4557
tr12-0255	0.5420	0.5546	0.6639	0.5462	0.4748	0.5630	0.5924	0.6329	0.5654
tr12-0441	0.6891	0.5756	0.5840	0.5168	0.5378	0.5630	0.5798	0.6498	0.5274
tr13-0170	0.6008	0.5630	0.5924	0.5546	0.6261	0.6471	0.5336	0.5612	0.4430
tr13-0508	0.4538	0.5084	0.6555	0.6008	0.5588	0.6261	0.6092	0.5654	0.5443
tr13-0653	0.6807	0.6303	0.5210	0.5462	0.5630	0.5336	0.6008	0.5359	0.5063
Average	0.5938	0.5854	0.5935	0.5424	0.5273	0.5536	0.5609	0.5510	0.5165