Research of Classification Algorithm Based on Local Coordination

Author

Jia, Liyuan ; Li, Lei ; Huang, Li

Author_Institution

Dept. of Comput. Sci., Hunan City Univ., Yiyang, China

Volume

2

fYear

2010

fDate

26-28 Aug. 2010

Firstpage

303

Lastpage

306

Abstract

Most of graph-based methods for semi-supervised learning are transductive, giving predictions for only the unlabeled data in the training set, and not for an arbitrary test point. SLC (Semi-supervised Local Linear Coordinate), which is based on LLC (Local Linear Coordinate) is present here as an inductive method. The mixture of factor analyzers is used to model the raw data set, and the label smoothness over the graph is enforced by local approximation. At last, smooth nonlinear projection is achieved by local affine transformation. Experiment shows the superiority of our proposed method in comparison to others.

Keywords

approximation theory; graph theory; learning (artificial intelligence); pattern classification; transforms; classification algorithm; graph-based methods; local affine transformation; local approximation; semi-supervised learning; semi-supervised local linear coordinate; smooth nonlinear projection; Approximation methods; Classification algorithms; Data models; Information processing; Machine learning; Manifolds; Training; local linear coordinate; manifold learning; mixture of factor analyzers; semi-supervised classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2010 2nd International Conference on

Conference_Location

Nanjing, Jiangsu

Print_ISBN

978-1-4244-7869-9

Type

conf

DOI

10.1109/IHMSC.2010.175

Filename

5591018