Clifford Manifold Learning for Nonlinear Dimensionality Reduction  

作　　者：CAO Wenming CAO Wenming 

机构地区：[1]School of Information Engineering, Shenzhen University, Shenzhen 518060, China [1]School of Information Engineering, Shenzhen University, Shenzhen 518060, China

出　　处：《Chinese Journal of Electronics》2009年第4期650-654,共5页电子学报（英文版）

基　　金：This work is supported by the National Natural Science Foundation of China （No.60576055）, and Pre-Research and Defense Fund of China （No.9140C80002080C80）.

摘　　要：The methods of manifold dimensionality reduction proposed recently are just for single dimensional data and signals. So this paper proposed the general frame- work of Clifford manifold learning, and solved the prob- lem of relationship between different dimensional signal and data using eigenmapping in local coordinate. We also mentioned the nonlinear dimensionality reduction analyses method based on Clifford algebra, and established the ho- mogeneous analyses model for Clifford nonlinear manifold with hybrid dimensional signals. The experiment and com- parison proved the efficiency of our method for nonlinear dimensionality reduction of hybrid dimensional signals.

关 键 词：Clifford algebra Manifold learning Hy- brid dimensional signal Dimensionality reduction. 

分 类 号：O151.24[理学—数学] TP18[理学—基础数学]