An Improved Dempster-Shafer Algorithm for Resolving the Conflicting Evidences

Dempster-Shafer evidence theory provides a useful computational scheme for integrating uncertainty information from multiple sources in artificial intelligence systems. Therefore, it has been successfully applied in data fusion and pattern recognition. However, it also has some shortcomings. D-S evidence combination can not proceed if the evidence totally collides with each other. In addition, the combination result can not be kept according to the real condition when the evidence seriously conflicts. To solve this problem, this paper presents an improved D-S algorithm, which verifies and modifies the conflicting evidences. Experiments show that this method improves the reliability and rationality of the combination results. Although evidences highly conflict with one another, the combination result still satisfies the practical situation.