Table 4. Comparison of different SSVMS with SVM and LSSVM in different data sets.
Data sets Performance SSVM |
Banknote = 1372*4 | EEG = 1500*20 | QSAR = 1055*41 | Crowdsourced = 2000*28 |
Diabetic = 1151*20 |
---|---|---|---|---|---|
SVM | 90.690 (336.294) |
62.330 (485.430) |
85.010 (1863.9) |
92.900 (5655.1) |
69.750 (3531.0) |
LSSVM | 87.670 (353.127) |
61.470 (0.912) |
82.200 (3.134) |
92.100 (2.108) |
63.330 (1.348) |
Sigmoid-SSVM | 88.484 (353.127) |
61.000 (283.030) |
83.212 (279.697) |
89.850 (853.942) |
57.863 (254.657) |
P2-SSVM | 88.484 (299.069) |
61.733 (243.039) |
83.318 (280.922) |
91.150 (879.012) |
60.016 (256.051) |
P4-SSVM | 88.776 (297.164) |
62.000 (273.971) |
83.697 (285.665) |
91.350 (927.105) |
60.035 (255.993) |
T3-SSVM | 88.921 (294.272) |
62.133 (244.845) |
83.697 (342.235) |
91.250 (899.858) |
64.987 (256.087) |
T5-SSVM | 88.994 (298.486) |
62.133 (245.261) |
83.886 (311.801) |
91.350 (939.453) |
64.992 (256.884) |
Padé22-SSVM | 89.723 (295.753) |
62.333 (250.239) |
84.171 (396.057) |
92.200 (852.956) |
65.682 (256.969) |
Padé33-SSVM |
89.923
(276.532) |
62.415
(228.632) |
84.435
(280.362) |
92.200
(836.634) |
68.215
(250.351) |