Table 6.

Classification comparison

Existing approaches	Feature #	Run #	D-66	D-160	D-255
DWT + SOM (Chaplot et al. 2006)	4761	5	94.00	93.17	91.65
DWT + SVM (Chaplot et al. 2006)	4761	5	96.15	95.38	94.05
DWT + SVM + RBF (Chaplot et al. 2006)	4761	5	98.00	97.33	96.18
DWT + SVM + POLY (Chaplot et al. 2006)	4761	5	98.00	97.15	96.37
DWT + PCA + KNN (El-Dahshan et al. 2010)	7	5	98.00	97.54	96.79
DWT + PCA + FP-ANN (El-Dahshan et al. 2010)	7	5	97.00	96.98	95.29
DWT + PCA + SCG-FNN (Dong et al. 2011)	19	5	100.00	99.27	98.82
DWT + PCA + SVM (Zhang and Wu 2012)	19	5	96.01	95.00	94.29
DWT + PCA + SVM + RBF (Zhang and Wu 2012)	19	5	100.00	99.38	98.82
DWT + PCA + SVM + IPOL (Zhang and Wu 2012)	19	5	100.00	98.12	97.73
DWT + PCA + SVM + HPOL (Zhang and Wu 2012)	19	5	98.34	96.88	95.61
RT + PCA + LS-SVM (Das et al. 2013)	9	5	100.00	100.00	99.39
DWT + SE + SWP + PNN (Saritha et al. 2013)	3	5	100.00	99.88	98.90
PCNN + DWT + PCA + BPNN (El-Dahshan et al. 2014)	7	10	100.00	98.88	98.24
SWT + PCA + IABAP-FNN (Wang et al. 2015a)	7	10	100.00	99.44	99.18
SWT + PCA + ABC-SPSO-FNN (Wang et al. 2015a)	7	10	100.00	99.75	99.02
WE + HMI + GEPSVM (Zhang et al. 2015d)	14	10	100.00	99.56	98.63

Proposed approach	Feature #	Run #	D-66	D-160	D-255
WPTE + FSVM	16	10	100.00	100.00	99.49

The italic represents the highest accuracy among all algorithms