A modification of the certainty measure to handle subnormal distributions
Authors:
Ronald R. Yager
Affiliation:
Machine Intelligence Institute, Iona College, New Rochelle, NY 10801, USA
Abstract:
We suggest a modification of the definition of certainty associated with a possibility distribution so that we are assured that the certainty measure is always less than or equal to the possibility measure even in the face of subnormal distributions. We note that when the possibility distribution is normal this new definition reduces to the original definition of certainty. We use this new measure along with the definition of possibility measure to obtain measures of belief and plausibility in the fact of fuzzy information in the Mathematical Theory of Evidence.