Journal of Shanghai Jiaotong University ›› 2012, Vol. 46 ›› Issue (04): 636-641.

• Automation Technique, Computer Technology • Previous Articles     Next Articles

Improved Combination Rule of Evidence Based on Pignistic Probability Distance

 XIAO  Jian-Yu-1, 2 , TONG  Min-Ming-1, ZHU  Chang-Jie-2, FAN  Qi-2   

  1. (1. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou 221116, Jiangsu, China; 2. School of Computer Science and Technology, Huaibei Normal University, Huaibei 235000, Jiangsu, China)
  • Received:2011-04-18 Online:2012-04-28 Published:2012-04-28

Abstract: To remedy the shortcomings in the recognition of evidence conflict by using conventional conflict coefficient, this paper discussed the application conditions of Dempster’s combination rule by traditional conflict quantification standard and the distance among probability assignment functions transformed by pignistic. Furthermore, the paper presented an improved method of evidence combination. In this method, the conflict level is represented by the pignistic probability distance of evidence, after that the conflict level is transformed into similarity level and the support degree of evidence is obtained. In addition, the weight coefficients are determined. Finally, the basic probability assignments adjusted by weight coefficient are fused by Dempster’s combination rule. The numerical examples prove that the modified method can resolve the evidence conflict efficiently and has fast convergence speed.

Key words:  evidence theory, combination rule, evidence conflict, pignistic probability distance

CLC Number: