Measuring Total Uncertainty in Dempster-Shafer Theory: A Novel Approach

Abstract A novel approach to measuring uncertainty and uncertainty-based information in Dempster-Shafer theory is proposed (independently also proposed by Maeda et al. 1993). It is shown that the proposed measure of total uncertainty in Dempster-Shafer theory is both additive and subadditive, has a desired range, and collapses correctly to either the Shannon entropy or the Hartley measure of uncertainty for special probability assignment functions. The paper is restricted, for the sake of simplicity, to finite sets. INDEX TERMS: UncertaintyDempster-Shafer theorymaximum entropy principle. Notes This research workwas spported in part by the NSF GrantNo. IST-90 15675.