• DocumentCode
    510282
  • Title

    A Modified Dempster-Shafer Combination Rule Based on Evidence Ullage

  • Author

    Ding, Yi-jie ; Wang, She-liang ; Zhao Xin-dong ; Lv Miao ; Yang Xi-qin

  • Author_Institution
    Sch. of Civil Eng., Xi´´an Univ. of Archit. & Technol., Xi´´an, China
  • Volume
    3
  • fYear
    2009
  • fDate
    7-8 Nov. 2009
  • Firstpage
    387
  • Lastpage
    391
  • Abstract
    Dempster-Shafer evidence theory is a powerful reasoning method in handling uncertain problems. Since Zadeh highlighted the counter-intuitive behaviour of Dempser´s combination rule, a lot of alternative rules have been proposed. However, most of these rules are merely the ¿patching¿ of the Dempser´s rule and lack of adequate consideration for evidence compatibility. In this paper, a modified combination rule is proposed based on the concept of evidence ullage. The suggested rule is a general extension of Dempster´s combination rule and aims at solving the counter-intuitive behaviour of evidence theory caused by evidence conflict and evidence compatibility. The parameters of evidence ullage and evidence tendency factor in the modified rule are also discussed and given. Calculated results show that the new rule can resolve the conflict and compatibility problems effectively.
  • Keywords
    uncertainty handling; Dempster-Shafer evidence theory; counter-intuitive behaviour; evidence ullage; reasoning method; uncertain handling problems; Artificial intelligence; Civil engineering; Computational intelligence; Computer architecture; Decision making; Electronic mail; Equations; Image processing; Signal detection; Uncertainty; Dempster-Shafer Evidence Theory; combination rule; evidence compatibility; evidence conflict; evidence ullage;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Artificial Intelligence and Computational Intelligence, 2009. AICI '09. International Conference on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-3835-8
  • Electronic_ISBN
    978-0-7695-3816-7
  • Type

    conf

  • DOI
    10.1109/AICI.2009.117
  • Filename
    5376711