Title :
Using discriminant analysis for multi-class classification
Author :
Li, Tao ; Zhu, Shenghuo ; Ogihara, Mitsunori
Author_Institution :
Dept. of Comput. Sci., Rochester Univ., NY, USA
Abstract :
Discriminant analysis is known to learn discriminative feature transformations. We study its use in multiclass classification problems. The performance is tested on a large collection of benchmark datasets.
Keywords :
computational complexity; learning (artificial intelligence); pattern classification; pattern recognition; statistical databases; support vector machines; SVM; benchmark dataset collection; discriminant analysis; discriminative feature transformation; machine learning problem; multiclass classification; support vector machine; Benchmark testing; Character generation; Chromium; Computer science; Covariance matrix; Face recognition; Linear discriminant analysis; Machine learning; Scattering; Support vector machines;
Conference_Titel :
Data Mining, 2003. ICDM 2003. Third IEEE International Conference on
Print_ISBN :
0-7695-1978-4
DOI :
10.1109/ICDM.2003.1250984
