Nonparametric kernel density estimation and its computational aspects 1st ed.

Author

• Gramacki, Artur

Additional Author(s)

-

Publisher

Cham: Springer International Publishing, 2018

Language

English

ISBN

9783319716886

Series

Studies in Big Data 37

Subject(s)

• ARTIFICIAL INTELLIGENCE
• DIGITAL FILTERS (MATHEMATICS)
• KERNEL FUNCTIONS

Notes

. .

Abstract

This book describes computational problems related to kernel density estimation (KDE) – one of the most important and widely used data smoothing techniques. A very detailed description of novel FFT-based algorithms for both KDE computations and bandwidth selection are presented. The theory of KDE appears to have matured and is now well developed and understood. However, there is not much progress observed in terms of performance improvements. This book is an attempt to remedy this. The book primarily addresses researchers and advanced graduate or postgraduate students who are interested in KDE and its computational aspects. The book contains both some background and much more sophisticated material, hence also more experienced researchers in the KDE area may find it interesting. The presented material is richly illustrated with many numerical examples using both artificial and real datasets. Also, a number of practical applications related to KDE are presented.

Physical Dimension

Number of Page(s)

1 online resource (xxix, 176 p.)

Dimension

-

Other Desc.

ill.

Summary / Review / Table of Content

Introduction.-
Nonparametric density estimation.-
Kernel density estimation .-
Bandwidth selectors for kernel density estimation.-
FFT-based algorithms for kernel density estimation and band-width selection.-
FPGA-based implementation of a bandwidth selection algorithm.-
Selected applications related to kernel density estimation.-
Conclusion and further research.

Exemplar(s)

#	Accession No.	Call Number	Location	Status
1.	00691/20	515.9 Gra N	Online !	Available