Research on conflicting evidences combination strategy
Xin Guan, Xiao Yi, You He · 2008
D-S evidence theory is a useful method for dealing with uncertainty problems. It can combine evidences from different evidence sources. Therefore, it has been successfully applied in data fusion and target recognition. However, the multiplying strategy in D-S evidence theory results in an inherent drawback when dealing with conflicting evidences. In order to efficiently combine highly conflicting evidences, a new D-S evidence combination strategy based on evidence classification and weight coefficients is put forward. In this method, firstly, evidences are classified into several subsets based on the principle that the evidences in each subset are consistent with each other. Secondly, a new method of determining weight coefficients is proposed. At last, all the combination results are integrated with weight rule to form the final result. Experiment results show that this method can process conflicting evidences more effectively compared with other existing methods.