Robust Semi-Supervised Manifold Learning Algorithm for Classification

Table 3

The classification rates on HAND_SHAPE data set for the different proportions (Training : Testing).

Training : Testing	5 : 2	3 : 2	1 : 1	2 : 3	2 : 5
LLE + NFL	0.8520	0.8200	0.8594	0.8552	0.8160
SS-LLE	0.9320	0.9057	0.8777	0.8562	0.8360
RSSML-LLE	0.9480	0.9343	0.9280	0.9095	0.8840
RLLPE + NFL	0.8980	0.8443	0.8411	0.8324	0.8128
SS-RLLPE	0.9300	0.8757	0.8617	0.8590	0.8392
RSSML-RLLPE	0.9460	0.9243	0.9326	0.8876	0.8744
LTSA + NFL	0.4200	0.3285	0.3657	0.3657	0.3200
SS-LTSA	0.8580	0.7829	0.8160	0.7951	0.7432
RSSML-LTSA	0.8640	0.7920	0.8223	0.8038	0.7488