Reasoning with Justifiable Exceptions in EL_bot Contextualized Knowledge Repositories

The Contextualized Knowledge Repository (CKR) framework has been proposed as a description logics-based approach for contextualization of knowledge, a well-known area of study in AI. The CKR knowledge bases are structured in two layers: a global context contains context-independent knowledge and contextual structure, while a set of local contexts hold specific knowledge bases. In practical uses of CKR, it is often desirable that global knowledge can be “overridden” at the local level, that is to recognize local pieces of knowledge that do not need to satisfy the general axiom. By targeting this need, in our recent works we presented an extension of CKR with global defeasible axioms, which apply local instances unless an exception for overriding exists; such an exception, however, requires that justification is provable from the knowledge base. In this paper we apply this framework to the basic description logic EL_ot. We provide a formalization of EL_ot CKRs with global defeasible axioms and study their semantic and computational properties. Moreover, we present a translation of CKRs to datalog programs under the answer set semantics for instance checking.