Comparison of Kernel Discriminant Coordinates Algorithms |
| |
Authors: | MirosŁaw KrzyŚko Łukasz Waszak |
| |
Affiliation: | Faculty of Mathematics and Computer Science, Adam Mickiewicz University, Umultowska, Poznań, Poland |
| |
Abstract: | The construction of kernel discriminant coordinates reduces to the solution of a generalized eigenvalue problem in which both matrices are nonnegative definite. Six different algorithms for solving that problem are described, and the performance of these algorithms is tested on 26 different datasets. The percentage of misclassifications using a linear discriminant function is noted, and the algorithms’ running times are ascertained. Classification is also performed in the space of classical discriminant coordinates. |
| |
Keywords: | Discriminant coordinates Generalized eigenvalue problem Kernel discriminant coordinates Moore–Penrose pseudoinverse Singular value decomposition. |
|