A Supervised Subspace Learning Algorithm: Supervised Neighborhood Preserving Embedding

Authors:
Xianhua Zeng;Siwei Luo
Affiliations:
School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China and School of Computer Science, China West Normal University, Sichuan 637002, China;School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
Venue:
ADMA '07 Proceedings of the 3rd international conference on Advanced Data Mining and Applications
Year:
2007

Citing 4
Cited 3

Think globally, fit locally: unsupervised learning of low dimensional manifolds

The Journal of Machine Learning Research
A Two-Stage Linear Discriminant Analysis via QR-Decomposition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Neighborhood Preserving Embedding

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
Supervised nonlinear dimensionality reduction for visualization and classification

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

Robust head pose estimation using supervised manifold learning

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part VI
Locally Minimizing Embedding and Globally Maximizing Variance: Unsupervised Linear Difference Projection for Dimensionality Reduction

Neural Processing Letters
Kernel Discriminant Embedding in face recognition

Journal of Visual Communication and Image Representation

Quantified Score

Hi-index	0.01

Visualization

Abstract

Neighborhood Preserving Embedding (NPE) is an unsupervised manifold learning algorithm with subspace learning characteristic. In fact, NPE is a linear approximation to Locally Linear Embedding (LLE). So it can provide an unsupervised subspace learning technique. In this paper, we proposed a new Supervised Neighborhood Preserving Embedding (SNPE) algorithm which can use the label or category information of training samples to better describe the intrinsic structure of original data in low-dimensional space. Furthermore, when a new unknown data needs to be processed, SNPE, as a supervised subspace learning technique, may be conducted in the original high-dimensional space. Several experiments on USPS digit database demonstrate the effectiveness of our algorithm.