DOI QR코드

DOI QR Code

Personalized EigenTrust with the Beta Distribution

  • Choi, Dae-Seon (Software Research Laboratory, ETRI) ;
  • Jin, Seung-Hun (Software Research Laboratory, ETRI) ;
  • Lee, Youn-Ho (Department of Information and Communication Engineering, Yeungnam University) ;
  • Park, Yong-Su (Department of Computer Science and Engineering, Hanyang University)
  • Received : 2009.08.25
  • Accepted : 2010.01.18
  • Published : 2010.04.30

Abstract

This letter presents an enhancement of EigenTrust. Using the beta distribution, local trust values can be more correctly evaluated. Simulation shows that the proposed scheme calculates the local trust more correctly by up to 8%. For personalization, the proposed scheme provides cumulative transitive values from the local trust to the global trust with mathematically guaranteed convergence.

Keywords

References

  1. S.D. Kamvar, M.T. Schlosser, and H. Garcia-Molina, "EigenTrust Algorithm for Reputation Management in P2P Networks," Proc. WWW Conf., 2003, pp. 640-651.
  2. U. Maurer, "Modelling a Public-Key Infrastructure," Proc. ESORICS, 1996, pp. 325-350.
  3. A. Gutscher, "A Trust Model for an Open, Decentralized Reputation System," Proc. IFIPTM, 2007, pp. 285-300.
  4. G. Shafer, A Mathematical Theory of Evidence, Princeton Univ. Press, 1976.
  5. K. Sentz and S. Ferson, Combination of Evidence in Dempster- Shafer Theory, SANDIA Tech. Report, SAND2002-0835, 2002.
  6. A. Josang, "Artificial Reasoning with Subjective Logic," Proc. AWCR, 1997, pp. 204-211.
  7. A. Gutscher, J. Heesen and O. Siemoneit, "Possibilities and Limitations of Modeling Trust and Reputation," Proc. CEUR Workshop, 2008.
  8. A. Josang and R. Ismail, "The Beta Reputation System," Proc. BCEC, 2002, pp. 324-337.
  9. Q. Lian et al., "Robust Incentives via Multilevel Tit-for-Tat," Concurrency and Computation: Practice & Experience, vol. 20, no. 2, 2008, pp. 167-178. https://doi.org/10.1002/cpe.1190