DMCA
Kernel Bandwidth Estimation for Nonparametric Modeling (2009)
Cached
Download Links
| Citations: | 4 - 1 self |
BibTeX
@MISC{Bors09kernelbandwidth,
author = {Adrian G. Bors and Nikolaos Nasios},
title = {Kernel Bandwidth Estimation for Nonparametric Modeling},
year = {2009}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Kernel density estimation is a nonparametric procedure for probability density modeling, which has found several applications in various fields. The smoothness and modeling ability of the functional approximation are controlled by the kernel bandwidth. In this paper, we describe a Bayesian estimation method for finding the bandwidth from a given data set. The proposed bandwidth estimation method is applied in three different computational-intelligence methods that rely on kernel density estimation: 1) scale space; 2) mean shift; and 3) quantum clustering. The third method is a novel approach that relies on the principles of quantum mechanics. This method is based on the analogy between data samples and quantum particles and uses the Schrödinger potential as a cost function. The proposed methodology is used for blind-source separation of modulated signals and for terrain segmentation based on topography information.
Keyphrases
nonparametric modeling kernel bandwidth estimation kernel density estimation functional approximation several application terrain segmentation scale space third method bandwidth estimation method probability density modeling novel approach nonparametric procedure schr dinger potential various field proposed methodology bayesian estimation method quantum mechanic mean shift kernel bandwidth different computational-intelligence method topography information quantum particle blind-source separation data set data sample cost function
