Capped -Norm Proximal Support Vector Machine

Table 16

Classification results on 6 vs 9 data with black block noise.

Dataset	GEPSVM	L1-GEPSVM	IGEPSVM	PCC	L1-NPSVM	LpNPSVM	GLpNPSVM	CPSVM
	Acc	Acc	Acc	Acc	Acc	Acc	Acc	Acc
Ori	92.69	94.15	97.21	82.05	65.29	98.01	98.27	98.01 (0.95)
B10	92.15	93.93	97.61	78.19	86.84	97.87	97.21	98.54 (0.85)
B20	84.74	93.43	96.45	78.95	81.97	95.92	96.84	97.24 (0.85)
B30	93.04	93.27	97.11	76.77	90.42	96.06	96.72	97.24 (0.7)
B40	93.47	93.04	96.48	79.77	66.84	95.30	96.34	97.24 (0.7)

Each bold value is the highest accuracy within each line.