An improved conflicting evidence combination approach based on a new supporting probability distance

Cholsok Yu, Jianhong Yang*, Debin Yang, Xianghong Ma, Hyonchun Min

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

To avoid counter-intuitive result of classical Dempster's combination rule when dealing with highly conflict information, many improved combination methods have been developed through modifying the basic probability assignments (BPAs) of bodies of evidence (BOEs) by using a certain measure of the degree of conflict or uncertain information, such as Jousselme's distance, the pignistic probability distance and the ambiguity measure. However, if BOEs contain some non-singleton elements and the differences among their BPAs are larger than 0.5, the current conflict measure methods have limitations in describing the interrelationship among the conflict BOEs and may even lead to wrong combination results. In order to solve this problem, a new distance function, which is called supporting probability distance, is proposed to characterize the differences among BOEs. With the new distance, the information of how much a focal element is supported by the other focal elements in BOEs can be given. Also, a new combination rule based on the supporting probability distance is proposed for the combination of the conflicting evidences. The credibility and the discounting factor of each BOE are generated by the supporting probability distance and the weighted BOEs are combined directly using Dempster's rules. Analytical results of numerical examples show that the new distance has a better capability of describing the interrelationships among BOEs, especially for the highly conflicting BOEs containing non-singleton elements and the proposed new combination method has better applicability and effectiveness compared with the existing methods.

Original languageEnglish
Pages (from-to)5139–5149
Number of pages11
JournalExpert Systems with Applications
Volume42
Issue number12
Early online date2 Mar 2015
DOIs
Publication statusPublished - 15 Jul 2015

Bibliographical note

Funding: National Natural Science Foundation of China (Nos. 50905013 and 51211130114), the Royal Society International Exchange Grant of United Kingdom (No. IE111065), the Fundamental Research Funds for the Central Universities special funds of China (No. FRF-TP-09-014A), and Beijing Higher Education Young Elite Teacher Project of China (No. YETP0364)

Keywords

  • evidence conflict
  • evidence theory
  • information fusion
  • Pignistic probability function
  • supporting probability function

Cite this

Yu, Cholsok ; Yang, Jianhong ; Yang, Debin ; Ma, Xianghong ; Min, Hyonchun. / An improved conflicting evidence combination approach based on a new supporting probability distance. In: Expert Systems with Applications. 2015 ; Vol. 42, No. 12. pp. 5139–5149.
@article{5e2ed0dac7594d76a7a34f9aee8b468f,
title = "An improved conflicting evidence combination approach based on a new supporting probability distance",
abstract = "To avoid counter-intuitive result of classical Dempster's combination rule when dealing with highly conflict information, many improved combination methods have been developed through modifying the basic probability assignments (BPAs) of bodies of evidence (BOEs) by using a certain measure of the degree of conflict or uncertain information, such as Jousselme's distance, the pignistic probability distance and the ambiguity measure. However, if BOEs contain some non-singleton elements and the differences among their BPAs are larger than 0.5, the current conflict measure methods have limitations in describing the interrelationship among the conflict BOEs and may even lead to wrong combination results. In order to solve this problem, a new distance function, which is called supporting probability distance, is proposed to characterize the differences among BOEs. With the new distance, the information of how much a focal element is supported by the other focal elements in BOEs can be given. Also, a new combination rule based on the supporting probability distance is proposed for the combination of the conflicting evidences. The credibility and the discounting factor of each BOE are generated by the supporting probability distance and the weighted BOEs are combined directly using Dempster's rules. Analytical results of numerical examples show that the new distance has a better capability of describing the interrelationships among BOEs, especially for the highly conflicting BOEs containing non-singleton elements and the proposed new combination method has better applicability and effectiveness compared with the existing methods.",
keywords = "evidence conflict, evidence theory, information fusion, Pignistic probability function, supporting probability function",
author = "Cholsok Yu and Jianhong Yang and Debin Yang and Xianghong Ma and Hyonchun Min",
note = "Funding: National Natural Science Foundation of China (Nos. 50905013 and 51211130114), the Royal Society International Exchange Grant of United Kingdom (No. IE111065), the Fundamental Research Funds for the Central Universities special funds of China (No. FRF-TP-09-014A), and Beijing Higher Education Young Elite Teacher Project of China (No. YETP0364)",
year = "2015",
month = "7",
day = "15",
doi = "10.1016/j.eswa.2015.02.038",
language = "English",
volume = "42",
pages = "5139–5149",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier",
number = "12",

}

An improved conflicting evidence combination approach based on a new supporting probability distance. / Yu, Cholsok; Yang, Jianhong; Yang, Debin; Ma, Xianghong; Min, Hyonchun.

In: Expert Systems with Applications, Vol. 42, No. 12, 15.07.2015, p. 5139–5149.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An improved conflicting evidence combination approach based on a new supporting probability distance

AU - Yu, Cholsok

AU - Yang, Jianhong

AU - Yang, Debin

AU - Ma, Xianghong

AU - Min, Hyonchun

N1 - Funding: National Natural Science Foundation of China (Nos. 50905013 and 51211130114), the Royal Society International Exchange Grant of United Kingdom (No. IE111065), the Fundamental Research Funds for the Central Universities special funds of China (No. FRF-TP-09-014A), and Beijing Higher Education Young Elite Teacher Project of China (No. YETP0364)

PY - 2015/7/15

Y1 - 2015/7/15

N2 - To avoid counter-intuitive result of classical Dempster's combination rule when dealing with highly conflict information, many improved combination methods have been developed through modifying the basic probability assignments (BPAs) of bodies of evidence (BOEs) by using a certain measure of the degree of conflict or uncertain information, such as Jousselme's distance, the pignistic probability distance and the ambiguity measure. However, if BOEs contain some non-singleton elements and the differences among their BPAs are larger than 0.5, the current conflict measure methods have limitations in describing the interrelationship among the conflict BOEs and may even lead to wrong combination results. In order to solve this problem, a new distance function, which is called supporting probability distance, is proposed to characterize the differences among BOEs. With the new distance, the information of how much a focal element is supported by the other focal elements in BOEs can be given. Also, a new combination rule based on the supporting probability distance is proposed for the combination of the conflicting evidences. The credibility and the discounting factor of each BOE are generated by the supporting probability distance and the weighted BOEs are combined directly using Dempster's rules. Analytical results of numerical examples show that the new distance has a better capability of describing the interrelationships among BOEs, especially for the highly conflicting BOEs containing non-singleton elements and the proposed new combination method has better applicability and effectiveness compared with the existing methods.

AB - To avoid counter-intuitive result of classical Dempster's combination rule when dealing with highly conflict information, many improved combination methods have been developed through modifying the basic probability assignments (BPAs) of bodies of evidence (BOEs) by using a certain measure of the degree of conflict or uncertain information, such as Jousselme's distance, the pignistic probability distance and the ambiguity measure. However, if BOEs contain some non-singleton elements and the differences among their BPAs are larger than 0.5, the current conflict measure methods have limitations in describing the interrelationship among the conflict BOEs and may even lead to wrong combination results. In order to solve this problem, a new distance function, which is called supporting probability distance, is proposed to characterize the differences among BOEs. With the new distance, the information of how much a focal element is supported by the other focal elements in BOEs can be given. Also, a new combination rule based on the supporting probability distance is proposed for the combination of the conflicting evidences. The credibility and the discounting factor of each BOE are generated by the supporting probability distance and the weighted BOEs are combined directly using Dempster's rules. Analytical results of numerical examples show that the new distance has a better capability of describing the interrelationships among BOEs, especially for the highly conflicting BOEs containing non-singleton elements and the proposed new combination method has better applicability and effectiveness compared with the existing methods.

KW - evidence conflict

KW - evidence theory

KW - information fusion

KW - Pignistic probability function

KW - supporting probability function

UR - http://www.scopus.com/inward/record.url?scp=84939939553&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2015.02.038

DO - 10.1016/j.eswa.2015.02.038

M3 - Article

AN - SCOPUS:84939939553

VL - 42

SP - 5139

EP - 5149

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 12

ER -