A Grey Wolf Optimizer for Modular Granular Neural Networks for Human Recognition

Table 28

Values of ORL database (test #2).

Method		Mean	Standard; deviation	Error standard deviation of the mean	Estimated; difference	-value	value	Degrees of freedom
Azami et al. [43]; (CGA + PCA)	5	95.91	0.409	0.18	−2.590	−8.96		8
Proposed method	5	98.50	0.500	0.22
Ch’Ng et al. [3] (PCA + LDA)	4	94.75	1.19	0.60	−3.750	−5.90	0.004	3
Proposed method	5	98.50	0.500	0.22
Sánchez et al. [38]; (FA)	5	98.30	0.447	0.20	−0.20	−0.67	0.523	8
Proposed method	5	98.50	0.500	0.22
Sánchez et al. [39]; (MGNNs + complexity)	5	97.59	1.71	0.76	−0.94	−1.15	0.314	4
Proposed method	5	98.50	0.500	0.22