Application of evidence Wasserstein distance algorithm in aspect of component identification

Abstract

The invention discloses application of an evidence Wasserstein distance algorithm in the aspect of component identification, the evidence Wasserstein distance is EWD, a new distance measurement algorithm is provided for the distance measurement problem in a D-S evidence theory, it is proved that a more reasonable conclusion can be obtained under the condition of high conflict, and the defect that the D-S evidence theory cannot obtain persuasion is overcome. And obtaining results under extreme conditions. Good characteristics such as symmetry, non-negativity, non-degradation and boundary are guaranteed, the innovation point of the invention lies in that the Wasserstein distance is combined with the Dane entropy for the first time, the evidence theory is introduced, and the introduction of the Wasserstein distance and the Deng entropy enables the evidence distance to have the capability of allocating high conflict factors during conflict measurement, so that the original attribute of the evidence is well reserved, and the method has the advantages of being high in robustness and high in robustness. Therefore, the possibility of generating confusion results is reduced, finally, practice and practical application can prove the superiority of the EWD method, and in a word, the proposed EWD effectively solves some problems when various evidences are combined under strong uncertainty.

Description

technical field [0001] The invention belongs to the technical field related to the evidence Wasserstein distance algorithm, and specifically relates to the application of the evidence Wasserstein distance algorithm in the aspect of component identification. Background technique [0002] In recent years, the uncertainty and incompleteness of relevant information has attracted more and more scholars' attention. Evidence theory originated from the evidence theory based on multi-valued mapping proposed by Dempster in 1967. Shafer transformed evidence theory from the uncertainty of propositions to the uncertainty relationship between sets, and was first applied to the field of artificial intelligence, with the ability to process and express uncertain information. Researchers have proposed a variety of methods to deal with information uncertainty, such as D number, Z number, entropy, etc. The related theories have been widely used in basic theory and practical applications. The ...

AI Technical Summary

Problems solved by technology

However, in cases of vast and outright conflict between various sources of evidence, the conclusions drawn by the Dempster-Shafer (D-S) theory of evidence often contradict common sense

Images