Dimension selection for feature selection and dimension reduction with principal and independent component analysis. Export

@article{citeulike:1034805, abstract = {This letter is concerned with the problem of selecting the best or most informative dimension for dimension reduction and feature extraction in high-dimensional data. The dimension of the data is reduced by principal component analysis; subsequent application of independent component analysis to the principal component scores determines the most nongaussian directions in the lower-dimensional space. A criterion for choosing the optimal dimension based on bias-adjusted skewness and kurtosis is proposed. This new dimension selector is applied to real data sets and compared to existing methods. Simulation studies for a range of densities show that the proposed method performs well and is more appropriate for nongaussian data than existing methods.}, address = {Department of Statistics, School of Mathematics, University of New South Wales, Sydney, NSW 2052 Australia, inge@maths.unsw.edu.au.}, author = {Koch, I. and Naito, K.}, citeulike-article-id = {1034805}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1246935}, citeulike-linkout-1 = {http://dx.doi.org/10.1162/neco.2007.19.2.513}, citeulike-linkout-2 = {http://view.ncbi.nlm.nih.gov/pubmed/17206873}, citeulike-linkout-3 = {http://www.hubmed.org/display.cgi?uids=17206873}, doi = {10.1162/neco.2007.19.2.513}, issn = {0899-7667}, journal = {Neural Comput}, keywords = {dimensionality, ica, pca}, month = {February}, number = {2}, pages = {513--545}, posted-at = {2007-01-11 03:55:54}, priority = {3}, title = {Dimension selection for feature selection and dimension reduction with principal and independent component analysis.}, url = {http://dx.doi.org/10.1162/neco.2007.19.2.513}, volume = {19}, year = {2007} }

TY - JOUR ID - citeulike:1034805 L3 - citeulike-article-id:1034805 N2 - This letter is concerned with the problem of selecting the best or most informative dimension for dimension reduction and feature extraction in high-dimensional data. The dimension of the data is reduced by principal component analysis; subsequent application of independent component analysis to the principal component scores determines the most nongaussian directions in the lower-dimensional space. A criterion for choosing the optimal dimension based on bias-adjusted skewness and kurtosis is proposed. This new dimension selector is applied to real data sets and compared to existing methods. Simulation studies for a range of densities show that the proposed method performs well and is more appropriate for nongaussian data than existing methods. IS - 2 JF - Neural Comput SN - 0899-7667 AD - Department of Statistics, School of Mathematics, University of New South Wales, Sydney, NSW 2052 Australia, inge@maths.unsw.edu.au. EP - 545 TI - Dimension selection for feature selection and dimension reduction with principal and independent component analysis. VL - 19 SP - 513 KW - dimensionality KW - ica KW - pca AU - Koch, I AU - Naito, K PY - 2007/02// UR - http://dx.doi.org/10.1162/neco.2007.19.2.513 ER -