Abstract
The paper addresses the problem of explaining some reasoning tasks associated with the DL-Lite Description Logic. Because of the simplicity of the language, standard concept level reasoning is quite easy, and the only contribution is an alternate, more accessible syntax, plus a novel focus on brevity of proofs. Of greater interest is the explanation of reasoning in finite models, which is motivated by the use of DL-Lite for database access. The fame of DL-Lite rests on its ability to answer efficiently conjunctive queries over KBs, and the paper makes three contributions in this regard: (1) a method for explaining why a value b was returned by a query; (2) a method for finding minimal explanations for why a conjunctive query is unsatisfiable; (3) the beginnings of a theory for explaining why a value b was not returned by a query.