<title>Efficient combination rule of evidence theory</title>

Abstract

D-S evidence theory is a useful method in dealing with uncertainty problems, but its application is limited because of the shortcomings of its combination rule. This paper presents an efficient combination rule, that is, the evidences' conflicting probability is distributed to every proposition according to its average supported degree. The new combination rule improves the reliability and rationality of combination results. Although evidences conflict one another highly, good combination results are also obtained.

Health Monitoring and Diagnosis of Equipment Based on Multi-sensor Fusion

“…Howev-er, the proportion of unknown set U is still large and the decision cannot be decided. Reference [18] simplifies the fusion rule on the basis of Reference [17], and it identifies the target A accurately when evidences are added to four pieces. Reference [20] introduces the credibility of evidence and correctly identifies the target A when the evidence is added to three pieces.…”

“…Reference [17] eliminates the "one vote veto" phenomenon, but it still cannot make a decision. The robustness of method in Reference [18] has been enhanced and the correct result can be obtained with four pieces of evidence. In Reference [20], the robustness is further strengthened and the correct result can be obtained with three pieces of evidence.…”

“…Reference [17] introduced a credibility of evidence and assigned the probabilities of conflicting evidences to various propositions, but the synthesis result still differed from common sense. In Reference [18], the probability of evidence conflict is weighted according to the average support degree to each proposition, and a relatively perfect synthetic result was obtained, but the complete set was ignored. Reference [19] proposed a method of averaging combination of evidence, which converged quickly but ignored the connection between evidences.…”

A New Preprocessing Method on Evidence Weight of Dempster Fusion Rule

“…Li Bi-cheng's method [8] , one kind of the modification of Yager's method, is a relatively mature one in the first kind of improvements. He thought that conflict information is not totally useless or uncertain, the belief on it should be reassigned.…”

Multisensor Data Fusion Based on Modified D-S Evidence Theory

“…The methods in the first category are to modify the D-S combination rule: Yager [6] proposed a conservative method assigning the conflict factors to the recognition framework. Li, et al [7] and Pan, et al [8] assigned conflict factor to all focal elements by a certain proportion, getting better results than that of Yager. The methods in the second category are to process the original evidence: Murphy, et al [9] assigned equal weight to each evidence by an average way.…”