Další formáty:
BibTeX
LaTeX
RIS
@inproceedings{566820,
author = {Forbelská, Marie},
address = {Prague},
booktitle = {COMPSTAT 2004 - Proceedings in Computational Statistics},
keywords = {linear and quadratic discriminant analysis; nonparametric discriminant analysis; kernel density estimation; product kernels; bandwidth choice},
language = {eng},
location = {Prague},
isbn = {3-7908-1554-3},
note = {poster presentation},
pages = {91-91},
publisher = {Physica-Verlag/Springer},
title = {A Study of Parametric and Nonparametric Kernel Density Discrimination (poster presentation)},
url = {http://www.compstat2004.cuni.cz/c04-poster-galery.htm},
year = {2004}
}
TY - JOUR
ID - 566820
AU - Forbelská, Marie
PY - 2004
TI - A Study of Parametric and Nonparametric Kernel Density Discrimination (poster presentation)
PB - Physica-Verlag/Springer
CY - Prague
SN - 3790815543
N1 - poster presentation
KW - linear and quadratic discriminant analysis
KW - nonparametric discriminant analysis
KW - kernel density estimation
KW - product kernels
KW - bandwidth choice
UR - http://www.compstat2004.cuni.cz/c04-poster-galery.htm
N2 - This paper compares the performance of parametric and nonparametric discrimination. The multivariate product Gaussian and polynomial kernels with various data-driven choices of the bandwidth are used for density estimators and this nonparametric approaches are compared with classical one by some real and simulated data. The Matlab software environment is used for preprocessing the data and to implement proposed classification methodology. A great attention is focused to the visualization of results.
ER -
FORBELSKÁ, Marie. A Study of Parametric and Nonparametric Kernel Density Discrimination (poster presentation). In \textit{COMPSTAT 2004 - Proceedings in Computational Statistics}. Prague: Physica-Verlag/Springer, 2004, s.~91-91. ISBN~3-7908-1554-3.
|
