An Uncertainty Measure for Interval-valued Evidences
An Uncertainty Measure for Interval-valued Evidences
复制标题
区间值证据的不确定性测度
DOI：
10.15837/ijccc.2017.5.2950
复制
发表时间：
2017-10-01
影响因子：
2.7
通讯作者：
Wang, S.
中科院分区：
文献类型：
--
作者：
Jiang, W.;Wang, S.
区间值信任结构(IBS)作为Dempster-Shafer证据理论中单值信任结构的扩展，逐渐被应用于许多领域。IBS将信任度赋予区间数而不是精确数，从而能够处理更复杂的不确定信息。然而，如何衡量IBS的不确定性仍然是一个悬而未决的问题。本文提出了一种新的基于邓氏熵的不确定度度量方法--UIV。此外，还证明了UIV满足一些理想的公理要求。文中给出了数值算例，通过与已有方法的比较，验证了该方法的有效性。
Interval-valued belief structure (IBS), as an extension of single-valued belief structures in Dempster-Shafer evidence theory, is gradually applied in many fields. An IBS assigns belief degrees to interval numbers rather than precise numbers, thereby it can handle more complex uncertain information. However, how to measure the uncertainty of an IBS is still an open issue. In this paper, a new method based on Deng entropy denoted as UIV is proposed to measure the uncertainty of the IBS. Moreover, it is proved that UIV meets some desirable axiomatic requirements. Numerical examples are shown in the paper to demonstrate the efficiency of UIV by comparing the proposed UIV with existing approaches.