earticle

논문검색

Research on Dynamic Trust Computing Method Based on Multi-Dimensional Evidence

초록

영어

From the perspective of the credibility of evaluating network main body, first consider to improving of evidence source on which trust computing is based, and propose the concept of multi-dimensional evidence. The multi-dimension refers to involvement of multiple types of evidence, and the multi-dimensional evidence, in this paper, involves mainly e-commerce business feedback evidence, online community business feedback evidence and network operation behavior evidence. On this basis, the evidence features can be incorporated into combination rule of evidence, design a new method to calculate the dynamic trust, and propose an improved D-S combination rule to synthesize multi-dimensional evidence, better to solve the problem of uncertainty of evidence. Experiments show it can effectively resist network fraud.

목차

Abstract
 1. Introduction
 2. D-S Evidence Theory
 3. Improvement of D-S Combination Rule
  3.1. Conflicting Intensity G
  3.2. Weighted Average to Belief Assignment Number
  3.3. Synthesize n-1 Times to n ma(A)
  3.4. Analyze the Effectiveness of G-Ge Combination Rule
 4. Experiment Analyses
  4.1. The Role Analysis of Evidences Weight
  4.2 Resistance to Collusion
  4.3 Resistance to Malicious Evaluation
 5. Conclusions
 Acknowledgements
 References

저자정보

  • Shi DeJia School of Computer and Information Engineering, Hunan University of Commerce, Changsha 410205, China,Information College, Central South University, Changsha 410000, China
  • Jiang WeiJin School of Computer and Information Engineering, Hunan University of Commerce, Changsha 410205, China, School of Computer, National University of Defense Technology, Changsha 410073, China
  • Zhang LianMei Electrical Engineering College, Wuhan University, Wuhan 430072, China

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.