Title :
Multi-classes Semi-supervised Learning on Riemannian Manifolds
Author :
Zhao, Zhong-Qiu ; Glotin, Herve ; Gao, Jun ; Wu, Xin-Dong
Author_Institution :
Comput. & Inf. Sch., Hefei Univ. of Technol., Hefei, China
Abstract :
A modified semi-supervised learning algorithm based on Riemannian manifolds is proposed, which extends the applicable field to the case of multi-classes problem. Furthermore, the modified version largely increases the learning speed, while attains the classification performance as satisfying as the original algorithm.
Keywords :
learning (artificial intelligence); Riemannian manifolds; multiclasses semisupervised learning algorithm; Computational intelligence; Educational programs; Laplace equations; Large scale integration; Semisupervised learning; Riemannian manifolds; learning speed; multi-classes problem; semi-supervised learning;
Conference_Titel :
Computational Intelligence and Natural Computing, 2009. CINC '09. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3645-3
DOI :
10.1109/CINC.2009.105
