Table 5 Model performance comparison across different datasets with 5-fold cross-validation results, showing test accuracy, recall, mean squared error (MSE), and computational efficiency. All accuracy/recall scores range [0,1], with higher values indicating better performance. MSE values reflect prediction error, and CPU times are reported in seconds.

Datasets	Models	Test Acc	Test Recall	Test MSE	CPU Time (s)
Iris	ASM	0.9667	0.9652	0.0163	0.02
	ASRM	0.9667	0.9652	0.0160	0.02
	GD	0.3933	0.3937	0.2645	2.04
Digits	ASM	0.9521	0.9525	0.0078	0.08
	ASRM	0.9521	0.9524	0.0078	0.07
	GD	0.8286	0.8269	0.0245	3.79
MNIST	ASM	0.8655	0.8640	0.0218	26.49
	ASRM	0.8657	0.8641	0.011	27.81
	GD	0.9755	0.9753	0.0040	101.29
Breast Cancer	ASM	0.9543	0.9415	0.0352	0.04
	ASRM	0.9543	0.9415	0.0352	0.02
	GD	0.5678	0.5591	0.3866	2.27
Wine	ASM	0.9944	0.9961	0.0035	0.03
	ASRM	0.9944	0.9961	0.0034	0.01
	GD	0.3637	0.3251	0.4175	1.89
Fashion MNIST	ASM	0.8228	0.8228	0.0291	24.47
	ASRM	0.8228	0.8229	0.0292	39.53
	GD	0.8875	0.8876	0.0162	100.96
SENTIMENT	ASM	0.8169	0.8131	6.3831	0.93
	ASRM	0.8208	0.8180	6.2410	3.81
	GD	0.8201	0.8155	0.1291	14.06
IFLYTEK	ASM	0.5141	0.3511	667.7762	2.37
	ASRM	0.5172	0.3635	666.9027	6.69
	GD	0.4195	0.2147	826.8232	15.80
TNEWS	ASM	0.4851	0.4909	18.0053	3.44
	ASRM	0.4860	0.4922	17.9442	13.77
	GD	0.4704	0.4530	18.3158	39.70
SHOPPING	ASM	0.7706	0.7399	5.2024	1.64
	ASRM	0.7723	0.7428	5.1462	12.30
	GD	0.7811	0.7495	4.7133	42.35

Quick links

Search