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

Table 5

The results of the 80% training and 20% testing experiment for six datasets.

Dataset	Algorithm	SSE	Training CA (%)	Testing CA (%)
Iris	MMPSO	0.31	100	100
	FWA	0.52	100	100
	KHA	21.28	99.59	100
	HS	18.00	98.33	96.67
	GA	96.00	90.00	90.00
Glass	MMPSO	47.56	78.36	62.79
	FWA	94.33	61.99	60.47
	KHA	41.21	58.79	58.14
	HS	355.85	70.12	72.09
	GA	544.00	57.89	67.44
Ionosphere	MMPSO	18.91	99.28	91.54
	FWA	25.28	95.71	90.14
	KHA	31.0	89.00	91.43
	HS	106.4	95.00	94.37
	GA	152	93.21	94.37
Breast cancer	MMPSO	47.00	99.46	97.85
	FWA	66.11	93.92	96.43
	KHA	-	-	-
	HS	126.37	-	100
	GA	172	-	98.57
Diabetes	MMPSO	166.21	81.27	79.87
	FWA	267.20	65.96	66.88
	KHA	-	-	-
	HS	856	-	77.27
	GA	1108	-	79.87
Thyroid	MMPSO	237.07	95.37	94.19
	FWA	749.11	93.21	93.82
	KHA	320.3	94.81	92.90
	HS	3146.4	93.06	92.78
	GA	3416.0	92.58	92.57