Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm

Jafar A. Alzubi, Ali Yaghoubi, Mehdi Gheisari, Yongrui Qin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Linear discriminant analysis is considered as current techniques in feature extraction so, LDA, by discriminant information which obtains in mapping space, does the classification act. When the classes' distribution is not normal, LDA, to perform classification, will face problem and will resulted the poor performance of criteria in performing the classification act. One of the proposed ways is the use of other measures, such as Chernoff's distance so, by using Chernoff's measure LDA has been spreading to its heterogeneous states and LDA in this state, in addition to use information among the medians, uses the information of the classes' Covariance matrices. By defining scattering matrix, based on Boundary and non- Boundary samples and using these matrices in Chernoff's criteria, the decrease of the classes' overlapping in the mapping space in as result, the rate of classification correctness increases. Using Boundary and non- Boundary samples in scattering matrices causes improvement over the result. In this article, we use a new discovering multi-stage Algorithm to choose Boundary and non- Boundary samples so, the results of the conducted experiments shows promising performance of the proposing method.
Original languageEnglish
Title of host publicationAlgorithms and Architectures for Parallel Processing
Subtitle of host publication18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II
EditorsJaideep Vaidya, Jin Li
Place of PublicationCham
PublisherSpringer Verlag
Pages130-144
Number of pages15
VolumeLNCS11335
ISBN (Electronic)9783030050542
ISBN (Print)9783030050535
DOIs
Publication statusPublished - 7 Dec 2018
Event18th International Conference on Algorithms and Architectures for Parallel Processing - Guangzhou, China
Duration: 15 Nov 201817 Nov 2018
Conference number: 18
http://nsclab.org/ica3pp2018/index.html (Link to Conference Website)

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Verlag
VolumeLNCS 11335
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Algorithms and Architectures for Parallel Processing
Abbreviated titleICA3PP 2018
CountryChina
CityGuangzhou
Period15/11/1817/11/18
Internet address

Fingerprint

Discriminant analysis
Scattering
Information use
Covariance matrix
Feature extraction
Experiments

Cite this

Alzubi, J. A., Yaghoubi, A., Gheisari, M., & Qin, Y. (2018). Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm. In J. Vaidya, & J. Li (Eds.), Algorithms and Architectures for Parallel Processing: 18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II (Vol. LNCS11335, pp. 130-144). (Lecture Notes in Computer Science; Vol. LNCS 11335). Cham: Springer Verlag. https://doi.org/10.1007%2F978-3-030-05054-2_10
Alzubi, Jafar A. ; Yaghoubi, Ali ; Gheisari, Mehdi ; Qin, Yongrui. / Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm. Algorithms and Architectures for Parallel Processing: 18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II. editor / Jaideep Vaidya ; Jin Li. Vol. LNCS11335 Cham : Springer Verlag, 2018. pp. 130-144 (Lecture Notes in Computer Science).
@inproceedings{59cfb90d80ce4d88832b99aae27b154d,
title = "Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm",
abstract = "Linear discriminant analysis is considered as current techniques in feature extraction so, LDA, by discriminant information which obtains in mapping space, does the classification act. When the classes' distribution is not normal, LDA, to perform classification, will face problem and will resulted the poor performance of criteria in performing the classification act. One of the proposed ways is the use of other measures, such as Chernoff's distance so, by using Chernoff's measure LDA has been spreading to its heterogeneous states and LDA in this state, in addition to use information among the medians, uses the information of the classes' Covariance matrices. By defining scattering matrix, based on Boundary and non- Boundary samples and using these matrices in Chernoff's criteria, the decrease of the classes' overlapping in the mapping space in as result, the rate of classification correctness increases. Using Boundary and non- Boundary samples in scattering matrices causes improvement over the result. In this article, we use a new discovering multi-stage Algorithm to choose Boundary and non- Boundary samples so, the results of the conducted experiments shows promising performance of the proposing method.",
keywords = "Boundary Pattern, CBP Algorithm, Chernoff Criterion, Linear Discriminant Analysis",
author = "Alzubi, {Jafar A.} and Ali Yaghoubi and Mehdi Gheisari and Yongrui Qin",
year = "2018",
month = "12",
day = "7",
doi = "10.1007{\%}2F978-3-030-05054-2_10",
language = "English",
isbn = "9783030050535",
volume = "LNCS11335",
series = "Lecture Notes in Computer Science",
publisher = "Springer Verlag",
pages = "130--144",
editor = "Jaideep Vaidya and Jin Li",
booktitle = "Algorithms and Architectures for Parallel Processing",

}

Alzubi, JA, Yaghoubi, A, Gheisari, M & Qin, Y 2018, Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm. in J Vaidya & J Li (eds), Algorithms and Architectures for Parallel Processing: 18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II. vol. LNCS11335, Lecture Notes in Computer Science, vol. LNCS 11335, Springer Verlag, Cham, pp. 130-144, 18th International Conference on Algorithms and Architectures for Parallel Processing, Guangzhou, China, 15/11/18. https://doi.org/10.1007%2F978-3-030-05054-2_10

Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm. / Alzubi, Jafar A.; Yaghoubi, Ali; Gheisari, Mehdi; Qin, Yongrui.

Algorithms and Architectures for Parallel Processing: 18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II. ed. / Jaideep Vaidya; Jin Li. Vol. LNCS11335 Cham : Springer Verlag, 2018. p. 130-144 (Lecture Notes in Computer Science; Vol. LNCS 11335).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm

AU - Alzubi, Jafar A.

AU - Yaghoubi, Ali

AU - Gheisari, Mehdi

AU - Qin, Yongrui

PY - 2018/12/7

Y1 - 2018/12/7

N2 - Linear discriminant analysis is considered as current techniques in feature extraction so, LDA, by discriminant information which obtains in mapping space, does the classification act. When the classes' distribution is not normal, LDA, to perform classification, will face problem and will resulted the poor performance of criteria in performing the classification act. One of the proposed ways is the use of other measures, such as Chernoff's distance so, by using Chernoff's measure LDA has been spreading to its heterogeneous states and LDA in this state, in addition to use information among the medians, uses the information of the classes' Covariance matrices. By defining scattering matrix, based on Boundary and non- Boundary samples and using these matrices in Chernoff's criteria, the decrease of the classes' overlapping in the mapping space in as result, the rate of classification correctness increases. Using Boundary and non- Boundary samples in scattering matrices causes improvement over the result. In this article, we use a new discovering multi-stage Algorithm to choose Boundary and non- Boundary samples so, the results of the conducted experiments shows promising performance of the proposing method.

AB - Linear discriminant analysis is considered as current techniques in feature extraction so, LDA, by discriminant information which obtains in mapping space, does the classification act. When the classes' distribution is not normal, LDA, to perform classification, will face problem and will resulted the poor performance of criteria in performing the classification act. One of the proposed ways is the use of other measures, such as Chernoff's distance so, by using Chernoff's measure LDA has been spreading to its heterogeneous states and LDA in this state, in addition to use information among the medians, uses the information of the classes' Covariance matrices. By defining scattering matrix, based on Boundary and non- Boundary samples and using these matrices in Chernoff's criteria, the decrease of the classes' overlapping in the mapping space in as result, the rate of classification correctness increases. Using Boundary and non- Boundary samples in scattering matrices causes improvement over the result. In this article, we use a new discovering multi-stage Algorithm to choose Boundary and non- Boundary samples so, the results of the conducted experiments shows promising performance of the proposing method.

KW - Boundary Pattern

KW - CBP Algorithm

KW - Chernoff Criterion

KW - Linear Discriminant Analysis

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85058660226&origin=resultslist&sort=plf-f&src=s&st1=Improve+Heteroscedastic+Discriminant+Analysis+by+Using+CBP+Algorithm&st2=&sid=feb792794c6ecae8a4cfc2d15ee17933&sot=b&sdt=b&sl=83&s=TITLE-ABS-KEY%28Improve+Heteroscedastic+Discriminant+Analysis+by+Using+CBP+Algorithm%29&relpos=0&citeCnt=1&searchTerm=

U2 - 10.1007%2F978-3-030-05054-2_10

DO - 10.1007%2F978-3-030-05054-2_10

M3 - Conference contribution

SN - 9783030050535

VL - LNCS11335

T3 - Lecture Notes in Computer Science

SP - 130

EP - 144

BT - Algorithms and Architectures for Parallel Processing

A2 - Vaidya, Jaideep

A2 - Li, Jin

PB - Springer Verlag

CY - Cham

ER -

Alzubi JA, Yaghoubi A, Gheisari M, Qin Y. Improve Heteroscedastic Discriminant Analysis by Using CBP Algorithm. In Vaidya J, Li J, editors, Algorithms and Architectures for Parallel Processing: 18th International Conference, ICA3PP 2018 Guangzhou, China, November 15-17, 2018 Proceedings, Part II. Vol. LNCS11335. Cham: Springer Verlag. 2018. p. 130-144. (Lecture Notes in Computer Science). https://doi.org/10.1007%2F978-3-030-05054-2_10