Author :
Cai, Deng ; He, Xiaofei ; Han, Jiawei
Abstract :
Recently the problem of dimensionality reduction (or, subspace learning) has received a lot of interests in many fields of information processing, including data mining, information retrieval, and pattern recognition. Some popular methods include principal component analysis (PCA), linear discriminant analysis (LDA) and locality preserving projection (LPP). However, a disadvantage of all these approaches is that the learned projective functions are linear combinations of all the original features, thus it is often difficult to interpret the results. In this paper, we propose a novel dimensionality reduction framework, called Unified Sparse Subspace Learning (USSL), for learning sparse projections. USSL casts the problem of learning the projective functions into a regression framework, which facilitates the use of different kinds of regularizes. By using a L1-norm regularizer (lasso), the sparse projections can be efficiently computed. Experimental results on real world classification and clustering problems demonstrate the effectiveness of our method.
Keywords :
data reduction; learning (artificial intelligence); regression analysis; data mining; dimensionality reduction; information processing; information retrieval; learned projective functions; linear discriminant analysis; locality preserving projection; pattern recognition; principal component analysis; spectral regression; unified sparse subspace learning; Clustering algorithms; Covariance matrix; Data mining; Helium; Information processing; Information retrieval; Linear discriminant analysis; Pattern recognition; Principal component analysis; Scattering;