• DocumentCode
    497609
  • Title

    Belief assignment on compound hypotheses within the framework of the Transferable Belief Model

  • Author

    Veremme, Alexandre ; Dupont, Daniel ; Lefevre, Eric ; Mercier, David

  • Author_Institution
    Dept. Ing. et Sci. du Vivant, HEI, Lille, France
  • fYear
    2009
  • fDate
    6-9 July 2009
  • Firstpage
    498
  • Lastpage
    505
  • Abstract
    Within the framework of evidence theory, the construction of mass functions remains an important problem which can considerably influence final results. Several propositions of construction are available in literature but generally, these methods do not consider the belief masses on compound hypotheses and do not exploit all the strengths of the theory. In this article, we propose two methods of grouping hypotheses allowing to reduce the number of focal elements during the construction of mass functions. The first one is based on a well-known approach in cartography and geoinformation fields: Jenks´ method. The second one is based on information criteria. We present and confront both methods. Results in a forensic entomology application are then presented.
  • Keywords
    belief networks; cartography; forensic science; geographic information systems; belief assignment; cartography; compound hypotheses; evidence theory; forensic entomology; geoinformation field; grouping hypotheses; mass functions; transferable belief model; Decision support systems; Displays; Forensics; Information processing; Knowledge management; Medical diagnostic imaging; Uncertainty; Jenks´ method; belief theory; compound hypotheses; evidence theory; forensic entomology; information criteria; mass function construction;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Fusion, 2009. FUSION '09. 12th International Conference on
  • Conference_Location
    Seattle, WA
  • Print_ISBN
    978-0-9824-4380-4
  • Type

    conf

  • Filename
    5203702