Share Email Print

Proceedings Paper

Comparison of discriminant analysis methods applied to stellar data classification
Author(s): Xi Wang; Fei Xing; Ping Guo
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In this study, five classifiers, namely quadratic discriminant analysis, linear discriminant analysis, regularlized discriminant analysis, leave-one-out covariance matrix estimate and Killback-Leibler information measure based method are considered for classification of stellar spectra data. Because stellar spectra data sets are severly ill-posed, we first adopt some feature selection method such as principal component analysis to reduce data dimensionality. The input of the classifiers are those selected features, and the cross-validation technique is used to optimize the regularization parameters. Experimental results show that in most cases, regularized classifiers are high classification rates than that of quadratic discriminant analysis, but parameter optimization is time consuming. From experiments of exhaustive searching regularization parameter, it is found that in some cases cross-validation method is not always good in the selection of models.

Paper Details

Date Published: 25 September 2003
PDF: 6 pages
Proc. SPIE 5286, Third International Symposium on Multispectral Image Processing and Pattern Recognition, (25 September 2003); doi: 10.1117/12.538644
Show Author Affiliations
Xi Wang, Beijing Normal Univ. (China)
Fei Xing, Beijing Normal Univ. (China)
Ping Guo, Beijing Normal Univ. (China)

Published in SPIE Proceedings Vol. 5286:
Third International Symposium on Multispectral Image Processing and Pattern Recognition
Hanqing Lu; Tianxu Zhang, Editor(s)

© SPIE. Terms of Use
Back to Top