System Z for Conditional Belief Bases with Positive and Negative Information

Authors

  • Meliha Sezgin TU Dortmund University
  • Gabriele Kern-Isberner

DOI:

https://doi.org/10.32473/flairs.v34i1.128398

Keywords:

conditional belief bases, positive and negative conditionals, system Z, consistency

Abstract

In non-monotonic reasoning, conditional belief bases mostly contain positive information in the form of standard conditionals. However, in practice we are often confronted with negative information, stating that a conditional does \emph{not} hold, i.e. we need a suitable approach for reasoning over belief bases $\Delta$ with positive and negative information. In this paper, we investigate the interaction of positive and negative information in a conditional belief base and establish a property for partitions of $\Delta$ that is equivalent to consistency. Based on this property, we develop a non-trivial extension of system Z for mixed conditional belief bases and provide an algorithm to compute this partition.

Downloads

Published

2021-04-18

How to Cite

Sezgin, M., & Kern-Isberner, G. (2021). System Z for Conditional Belief Bases with Positive and Negative Information. The International FLAIRS Conference Proceedings, 34. https://doi.org/10.32473/flairs.v34i1.128398

Issue

Section

Special Track: Uncertain Reasoning