In this paper, we consider computing the difference between two Horn theories. This problem may arise, for example, if we take care of a theory change in a knowledge base. In general, the difference of Horn theories is not Horn. Therefore, we consider Horn approximations of the difference in terms of Horn cores (i.e., weakest Horn theories included in the difference). We study the problem under the familiar representation of Horn theories by Horn CNFs, as well as under the recently proposed model-based representation in terms of the characteristic models. For all problems and representations, polynomial time algorithms or proofs of intractability for the propositional case are provided.
CITATION STYLE
Eiter, T., Ibaraki, T., & Makino, K. (1999). On the difference of horn theories (Extended abstract). In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1563, pp. 467–477). Springer Verlag. https://doi.org/10.1007/3-540-49116-3_44
Mendeley helps you to discover research relevant for your work.