Evidence combination based on credibility and non-specificity
This paper addresses the combination of unreliable evidence sources which provide uncertain information in the form of basic probability assignment (BPA) functions. We proposed a novel evidence combination rule based on credibility and non-specificity of belief functions. Following a review of all existing non-specificity measures in evidence theory, a non-specificity measure for evidence theory is discussed. It is claimed that the non-specificity degree of a BPA is related to its ability of pointing to one and only one element. Based on the difference between the largest belief grades and other belief grades, a non-specificity measure is defined. Properties of the proposed non-specificity measure are put forward and proved mathematically. Illustrative examples are employed to show the properties of non-specificity measure. After providing a procedure for the evaluation of evidence credibility, we propose a novel evidence combination rule. Numerical example and application in target identification are applied to demonstrate the performance of our proposed evidence combination rule.
KeywordsEvidence theory Belief function Non-specificity degree Credibility degree
- 6.Smets P (2000) Data fusion in the transferable belief model. In: Proceedings of the 3rd international conference on information fusion, Paris, France, pp PS21–PS33Google Scholar
- 10.Haenni R (2005) Shedding new light on Zadeh’s criticism of Dempster’s rule of combination. In: Proceedings of the eighth international conference on information fusion, Philadelphia, USA. IEEE, Piscataway, pp 879–884Google Scholar
- 13.Klein J, Colot O (2010) Automatic discounting rate computation using a dissent criterion. In: Proceedings of the workshop on the theory of belief functions, Brest, France, pp 1–6Google Scholar
- 28.Yager RR (1981) Measurement of properties of fuzzy sets and possibility distributions. In: Proceedings of third international seminar on fuzzy sets, Linz, pp 211–222Google Scholar
- 32.Song Y, Wang X, Lei L, Xue A (2014) Measurement of evidence conflict based on correlation coefficient. J Commun 35(5):95–100Google Scholar