Distances and kernels based on cumulative distribution functions

Hongjun Su; Hong Zhang

doi:10.1016/B978-0-12-802045-6.00036-3

Distances and kernels based on cumulative distribution functions

Hongjun Su, Hong Zhang

Armstrong Atlantic State University

Research output: Contribution to book or proceeding › Chapter › peer-review

5 Scopus citations

Abstract

Similarity and dissimilarity measures such as kernels and distances are key components of classification and clustering algorithms. We propose a novel technique to construct distances and kernel functions between probability distributions based on cumulative distribution functions. The proposed distance measures incorporate global discriminating information and can be computed efficiently.

Original language	English
Title of host publication	Emerging Trends in Image Processing, Computer Vision and Pattern Recognition
Publisher	Elsevier Inc.
Pages	551-559
Number of pages	9
ISBN (Electronic)	9780128020920
ISBN (Print)	9780128020456
DOIs	https://doi.org/10.1016/B978-0-12-802045-6.00036-3
State	Published - 2015

Keywords

Cumulative distribution function
Distance
Kernel
Similarity

Access to Document

10.1016/B978-0-12-802045-6.00036-3

Cite this

@inbook{207ddf57ab7e4ae8b6892321cb9962f6,

title = "Distances and kernels based on cumulative distribution functions",

abstract = "Similarity and dissimilarity measures such as kernels and distances are key components of classification and clustering algorithms. We propose a novel technique to construct distances and kernel functions between probability distributions based on cumulative distribution functions. The proposed distance measures incorporate global discriminating information and can be computed efficiently.",

keywords = "Cumulative distribution function, Distance, Kernel, Similarity",

author = "Hongjun Su and Hong Zhang",

year = "2015",

doi = "10.1016/B978-0-12-802045-6.00036-3",

language = "English",

isbn = "9780128020456",

pages = "551--559",

booktitle = "Emerging Trends in Image Processing, Computer Vision and Pattern Recognition",

publisher = "Elsevier Inc.",

address = "United States",

}

TY - CHAP

T1 - Distances and kernels based on cumulative distribution functions

AU - Su, Hongjun

AU - Zhang, Hong

PY - 2015

Y1 - 2015

N2 - Similarity and dissimilarity measures such as kernels and distances are key components of classification and clustering algorithms. We propose a novel technique to construct distances and kernel functions between probability distributions based on cumulative distribution functions. The proposed distance measures incorporate global discriminating information and can be computed efficiently.

AB - Similarity and dissimilarity measures such as kernels and distances are key components of classification and clustering algorithms. We propose a novel technique to construct distances and kernel functions between probability distributions based on cumulative distribution functions. The proposed distance measures incorporate global discriminating information and can be computed efficiently.

KW - Cumulative distribution function

KW - Distance

KW - Kernel

KW - Similarity

UR - http://www.scopus.com/inward/record.url?scp=84942776738&partnerID=8YFLogxK

U2 - 10.1016/B978-0-12-802045-6.00036-3

DO - 10.1016/B978-0-12-802045-6.00036-3

M3 - Chapter

AN - SCOPUS:84942776738

SN - 9780128020456

SP - 551

EP - 559

BT - Emerging Trends in Image Processing, Computer Vision and Pattern Recognition

PB - Elsevier Inc.

ER -

Distances and kernels based on cumulative distribution functions

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this