Improvement of D-S Fusion Algorithm under Evidence Conflict
Huihui Sun · Jisuanji fangzhen · 2013
Aiming at the situation that appeared paradox when using D-S evidence theory to deal with the high conflict evidences,under the framework of the D-S evidence theory,the deciding method of conflicts was given,and an improved mutual weight combination algorithm was proposed.This approach made use of the evidence distance to determine the mutual support degree,and eigenvector for the maximal eigenvalue of evidence support degree matrix was considered to be weight vector.Then evidence discount coefficient could be gained,and it was used to every evidence,which was combined by D-S rule,and this avoided the paradox of D-S combination rule under the information conflict.In order to test the validity of this algorithm,simulation was given.And the algorithm validates that it can effectively decrease the effects that the conflict makes to combination results.And this algorithm reduces the decision-making risk.