Name
Abstract | ||
|---|---|---|
. The nonlinear classifier is effective for many practical problems.We have already proposed a method for constructing a nonlinearclassifier using Legendre polynomials and have obtained good results onmany actual data. In this approach, a set of original features is firstextended to a large number of new features in a nonlinear fashion andthen some substantial features are chosen for the nonlinear classifier. Inthis study, we have improved the selection process in the second stage by... |
Year | DOI | Venue |
|---|---|---|
1998 | 10.1007/BFb0033279 | SSPR/SPR |
Keywords | Field | DocType |
feature selection,nonlinear classifier,legendre polynomial | Nonlinear system,Feature selection,Polynomial,Pattern recognition,Computer science,Algorithm,Greedy algorithm,Preprocessor,Artificial intelligence,Margin classifier,Classifier (linguistics),Quadratic classifier | Conference |
ISBN | Citations | PageRank |
3-540-64858-5 | 0 | 0.34 |
References | Authors | |
4 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| M. Sato | 1 | 1 | 1.40 |
| Mineichi Kudo | 2 | 927 | 116.09 |
| Jun Toyama | 3 | 130 | 19.87 |
| Masaru Shimbo | 4 | 179 | 33.02 |