Založeno v roce 2005 s podporou MŠMT ČR (projekt 1M0572)

Publikace

Flexible-hybrid Sequential Floating Search in Statistical Feature Selection.

Typ:
Článek v odborném periodiku
Autoři publikace:
Název periodika:
Lecture Notes in Computer Science
Rok:
2006
Číslo:
4109
Strany:
632-639
ISSN:
0302-9743
Anotace:
A flexible hybrid version of floating search methods for feature selection is developed. The main benefit of the proposed floating search hybridization is the possibility to deal flexibly with the quality-of-result vs. computational time trade-off and to enable wrapper based feature selection in problems of higher dimensionality than before. It is shown that it is possible to trade significant reduction of search time for often negligible decrease of the classification accuracy.
 
Copyright 2005 DAR XHTML CSS