Reasoning in Description Logics


tively used in the construction of knowledge-based applications in several domains (see for example Wright et al. 1993), like connguration and software engineering. Recently we have seen attempts to use them in the elds of machine learning (see for example Cohen and Hirsh 1994) and planning (see for example Devambu and Litman 1991, Weida and Litman 1992, Ar-tale and Franconi 1994, where the basic formalism is extended to deal with time and action). Related elds. Finally, there are close relationships between Description Logics and formalisms developed in other areas of computer science. In particular, there is a strict relationship with logics of programs, originally studied for program veriication and analysis (see Schild 1991, Schild 1994, De Giacomo and Lenzerini 1994a). As we said in Sections 5 and 6, this correspondence has provided new useful tools for studying description logics equipped with a very expressive concept language, as well as insights on the deenitional mechanisms of the TBox. Moreover, there are strong links with the eld of Databases and in particular with Semantic and Object-oriented Data Models (see for example Research in this direction can be signiicant not only for the exchange of techniques and results, but may have a strong impact on the development of tools for database speciication and design. by two parts: axioms about concepts in the TBox representing intensional knowledge, and assertions about individuals in the ABox representing ex-tensional knowledge. We arranged the presentation by considering diierent TBox-ABox settings, and looking at corresponding reasoning problems. However, research on Description Logics has made several other contributions , as the reader can nd in a series of workshops speciically dedicated to this subject (Nebel et al. 1991, MacGregor et al. 1992, Baader et al. 1994, Borgida et al. 1995). In the following we brieey mention some of the issues, which were outside the scope of this survey. Implementation issues. The implementation of DL-systems involves many other aspects besides the computational analysis of reasoning problems. Engineering of knowledge representation systems is illustrated for example in (Brachman 1992). More speciically, with regard to the implementation of reasoning procedures, the tableaux-based reasoning techniques which we have considered in this survey should be equipped with specialized strategies for rule selection and application. Another approach to the characterization of reasoning methods is to look at existing implementations and then try to characterize the computations done by the system , possibly using a non-standard semantics …


Citations per Year

542 Citations

Semantic Scholar estimates that this publication has 542 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Donini1997ReasoningID, title={Reasoning in Description Logics}, author={Francesco M. Donini and Maurizio Lenzerini and Daniele Nardi and Andrea Schaerf and Elia Weixelbaum and Gregg T. Vesonder and Karen E. Brown and Stephen R. Palmer and Jay I. Berman and Ron J. Brachman and Hector J. Levesque}, year={1997} }