Robust Semi-Supervised Manifold Learning Algorithm for Classification

Table 1

The classification rates of the 9 classification methods on three different noisy data sets (CMU PIE, HAND_SHAPE and HW-Alpha).

METHOD	CMU PIE	HAND_SHAPE	HW-Alpha
LLE + NFL	0.6088	0.8549	0.7030
SS-LLE	0.6613	0.8777	0.7607
RSSML-LLE	0.7763	0.9280	0.7792
RLLPE + NFL	0.6263	0.8411	0.7294
SS-RLLPE	0.6375	0.8617	0.7714
RSSML-RLLPE	0.6963	0.9326	0.7778
LTSA + NFL	0.4488	0.3657	0.3205
SS-LTSA	0.8300	0.8160	0.7500
RSSML-LTSA	0.8500	0.8183	0.7585