A Novel Multimean Particle Swarm Optimization Algorithm for Nonlinear Continuous Optimization: Application to Feed-Forward Neural Network Training

Table 3

The results of 5-fold cross validation experiment of the proposed MMPSO and PSO.

Dataset	Algorithm	SSE	Training CA (%)	Testing CA (%)
Lymphography	PSO	23.46	89.32	78.67
	MMPSO	13.87	91.49	83.67
Iris	PSO	6.72	91.00	90.67
	MMPSO	1.35	97.33	91.67
Wine	PSO	3.86	92.25	87.22
	MMPSO	0.39	99.58	95.56
Glass	PSO	83.60	66.35	55.84
	MMPSO	62.36	69.21	58.14
Shuttle-landing	PSO	2.07	99.51	94.12
	MMPSO	1.79	99.51	96.08
Ionosphere	PSO	24.77	93.07	85.35
	MMPSO	20.16	95.79	88.73
Balance-scale	PSO	58.21	90.20	86.56
	MMPSO	35.85	90.24	89.12
Breast cancer	PSO	73.80	98.38	95.14
	MMPSO	59.43	98.57	96.29
Diabetes	PSO	202.39	76.78	72.08
	MMPSO	182.54	77.55	78.05
Thyroid	PSO	528.55	92.95	92.83
	MMPSO	373.63	93.66	94.01