The Degree of Conflict Between Formulas in an Inconsistent Knowledge Base

Measuring inconsistency in a knowledge base has been increasingly recognized as a good starting point to better understand the inconsistency of that base. Most approaches to measuring inconsistency for knowledge bases focus on either the degree of inconsistency of a whole knowledge base or the responsibility of each formula of a knowledge base for the inconsistency in that base or both. In this paper, we propose an approach to measuring the degree of conflict between formulas, which allows us to have a more clear picture on the relation between formulas involved in inconsistency. Then we show that this measure can be explained well in the framework of Halpern-Pearl's causal model.