Drawbacks of Uncertainty Measures Based on the Pignistic Transformation

Recently, a measure of “total uncertainty” (TU) in Dempster–Shafer theory, based on the pignistic distribution called ambiguity measure (AM), have been modified. The resulting new measure has been simply referred as modified AM (MAM). In the literature, it has been shown that AM, in addition to showing some undesirable behaviors, has important drawbacks related to two essential properties for such measures: 1) subadditivity and 2) monotonicity. The MAM measure has been developed to solve the AM subadditivity problem, but this paper demonstrates that MAM suffers the same drawback as AM with respect to monotonicity. A measure of uncertainty that cannot meet the monotonicity requirement has an important drawback for its exploitation in operational contexts such as in analytics, information fusion, and decision support. This paper aims at identifying and discussing drawbacks of this type of measures (AM, MAM). Our main motivation is to insist upon the important requirement of monotonicity that a TU measure should possess to improve its potential of being used and trusted in applications. This discussion is due time since the monotonicity problem needs first to be solved to avoid building too high expectations for usefulness and potential exploitation of such measures in operational communities.

[1]  G. Klir,et al.  MEASURING TOTAL UNCERTAINTY IN DEMPSTER-SHAFER THEORY: A NOVEL APPROACH , 1994 .

[2]  G. Klir,et al.  MEASURES OF UNCERTAINTY AND INFORMATION BASED ON POSSIBILITY DISTRIBUTIONS , 1982 .

[3]  George J. Klir,et al.  Remarks on “Measuring Ambiguity in the Evidence Theory” , 2008, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[4]  Asghar Shahpari,et al.  Using Mutual Aggregate Uncertainty Measures in a Threat Assessment Problem Constructed by Dempster–Shafer Network , 2015, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[5]  Andrés R. Masegosa,et al.  Requirements for total uncertainty measures in Dempster–Shafer theory of evidence , 2008, Int. J. Gen. Syst..

[6]  Fred Richman,et al.  Calculating Maximum-Entropy Probability densities for Belief Functions , 1994, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[7]  Serafín Moral,et al.  A non-specificity measure for convex sets of probability distributions , 2000 .

[8]  Jean Dezert,et al.  Foundations for a new theory of plausible and paradoxical reasoning , 2002 .

[9]  Éloi Bossé,et al.  Measuring ambiguity in the evidence theory , 2006, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[10]  Joaquín Abellán.  Combining nonspecificity measures in Dempster–Shafer theory of evidence , 2011, Int. J. Gen. Syst..

[11]  Éloi Bossé,et al.  Reducing Algorithm Complexity for Computing an Aggregate Uncertainty Measure , 2007, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[12]  Serafín Moral,et al.  Difference of entropies as a non-specificity function on credal sets† , 2005, Int. J. Gen. Syst..

[13]  George J. Klir,et al.  Disaggregated total uncertainty measure for credal sets , 2006, Int. J. Gen. Syst..

[14]  Serafín Moral,et al.  An Algorithm to Compute the Upper Entropy for Order-2 Capacities , 2006, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[15]  Serafín Moral,et al.  Maximum of Entropy for Credal Sets , 2003, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[16]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[17]  Serafín Moral,et al.  Upper entropy of credal sets. Applications to credal classification , 2005, Int. J. Approx. Reason..

[18]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[19]  Alex M. Andrew,et al.  Uncertainty and Information: Foundations of Generalized Information Theory , 2006 .

[20]  D. Dubois,et al.  A NOTE ON MEASURES OF SPECIFICITY FOR FUZZY SETS , 1985 .

[21]  R. Hartley.  Transmission of information , 1928 .

[22]  Arthur P. Dempster,et al.  Upper and Lower Probabilities Induced by a Multivalued Mapping , 1967, Classic Works of the Dempster-Shafer Theory of Belief Functions.