Reasoning with Mass Distributions
Artificial Intelligence
2013-03-26 v1
Abstract
The concept of movable evidence masses that flow from supersets to subsets as specified by experts represents a suitable framework for reasoning under uncertainty. The mass flow is controlled by specialization matrices. New evidence is integrated into the frame of discernment by conditioning or revision (Dempster's rule of conditioning), for which special specialization matrices exist. Even some aspects of non-monotonic reasoning can be represented by certain specialization matrices.
Cite
@article{arxiv.1303.5726,
title = {Reasoning with Mass Distributions},
author = {Rudolf Kruse and Detlef Nauck and Frank Klawonn},
journal= {arXiv preprint arXiv:1303.5726},
year = {2013}
}
Comments
Appears in Proceedings of the Seventh Conference on Uncertainty in Artificial Intelligence (UAI1991)