Learning sparse features for classification by mixture models

Authors:
Weixiang Liu;Nanning Zheng
Affiliations:
Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an, Shaanxi Province 710049, PR China;Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an, Shaanxi Province 710049, PR China
Venue:
Pattern Recognition Letters
Year:
2004

Citing 1
Cited 6

Statistical Pattern Recognition: A Review

IEEE Transactions on Pattern Analysis and Machine Intelligence

SVD based initialization: A head start for nonnegative matrix factorization

Pattern Recognition
Sparse p-norm Nonnegative Matrix Factorization for clustering gene expression data

International Journal of Data Mining and Bioinformatics
Sparse p-norm Nonnegative Matrix Factorization for clustering gene expression data

International Journal of Data Mining and Bioinformatics
Robust automatic data decomposition using a modified sparse NMF

MIRAGE'07 Proceedings of the 3rd international conference on Computer vision/computer graphics collaboration techniques
Semantic feature extraction for brain CT image clustering using nonnegative matrix factorization

ICMB'08 Proceedings of the 1st international conference on Medical biometrics
Two-dimensional non-negative matrix factorization for face representation and recognition

AMFG'05 Proceedings of the Second international conference on Analysis and Modelling of Faces and Gestures

Quantified Score

Hi-index	0.10

Visualization

Abstract

Non-negative matrix factorization (NMF) can discover sparse features for classification via mixture models and the sparseness of features controls the learning rate of the basis function parameters. But the original NMF in which the basis vectors are unit ones in L1 norm, does not increase the sparseness of learned features. This paper generalizes NMF to Lp-NMF where the basis vectors are unit ones in Lp norm. Experiments demonstrate how p affects the sparseness of learned features and the final classification accuracy. And the results show that L2-NMF is superior one for practical implementation.