Reasoning with Mass Distributions
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
Text
Kruse et al. "Reasoning with Mass Distributions." Conference on Uncertainty in Artificial Intelligence, 1991.Markdown
[Kruse et al. "Reasoning with Mass Distributions." Conference on Uncertainty in Artificial Intelligence, 1991.](https://mlanthology.org/uai/1991/kruse1991uai-reasoning/)BibTeX
@inproceedings{kruse1991uai-reasoning,
title = {{Reasoning with Mass Distributions}},
author = {Kruse, Rudolf and Nauck, Detlef and Klawonn, Frank},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {1991},
url = {https://mlanthology.org/uai/1991/kruse1991uai-reasoning/}
}