• DocumentCode
    2298870
  • Title

    A Dynamic Trust Measurement Based on Characteristics of Evaluation Indexes

  • Author

    Dong Yu-xin ; Yin Gui-sheng ; Teng Xu-yang

  • Author_Institution
    Coll. of Comput. Sci. & Technol., Harbin Eng. Univ., Harbin, China
  • fYear
    2010
  • fDate
    1-2 Nov. 2010
  • Firstpage
    182
  • Lastpage
    187
  • Abstract
    To deal with the features of a large variety of evaluation indexes in current Internetware trust measurement as well as the problem of how to select evaluation indexes, a dynamic trust evaluation method combining historical instance data and characteristics of evaluation indexes is developed in this paper. In developing this method, an evaluation index hierarchy is constructed based on historical instance data of Internetware. After categorizing indexes according to their characteristics, the trust value of each index is measured and calculated. A weight for each index is determined accordingly by entropy of information. The overall trust value is gained by combining trust values and weights. The similar components set is achieved through overall cluster analysis. The experimental result indicates that the proposed evaluation method has the function of evaluating the trust level of components effectively and reasonably, and provides scientific evidences for the process of the trust transfer and combination.
  • Keywords
    Internet; entropy; security of data; Internetware trust measurement; cluster analysis; dynamic trust evaluation method; dynamic trust measurement; evaluation index; historical instance data; information entropy; trust transfer; Attenuation; Current measurement; Entropy; Indexes; Information entropy; Probabilistic logic; Stability criteria; Internetware; cluster analysis; evaluation indexes; information entropy;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Internet Computing for Science and Engineering (ICICSE), 2010 Fifth International Conference on
  • Conference_Location
    Heilongjiang
  • Print_ISBN
    978-1-4244-9954-0
  • Type

    conf

  • DOI
    10.1109/ICICSE.2010.42
  • Filename
    6076566