Motivation

Knowledge Representation is the field of Artificial Intelligence which focusses on the design of formalisms that are both epistemologically and computationally adequate for expressing knowledge on a particular domain. One of the main research lines of the field has been concerned with the idea that the knowledge should be represented by characterizing classes of objects and the relationship among them. The organization of the classes used to describe a domain of interest is based on a hierarchical structure, which not only provides an effective and compact representation of information, but also allows for performing the relevant reasoning tasks in a computationally effective way.

The above principle was underlying the development of the first frame systems and semantic networks. However, such systems were in general not formally defined and the associated reasoning tools were strongly dependent on the implementation strategies. A fundamental step towards a logic-based characterization of such systems has been accomplished through the work on the KL-ONE system, which collected many of the ideas stemming from earlier semantic networks and frame-based systems, and provided a logical basis for interpreting objects, classes (or concepts), and relationships (or links, or roles) between them. The first goal of such a logical reconstruction was the precise characterization of the set of constructs used to build class and link expressions. The second goal was to provide reasoning procedures that are sound and complete with respect to the semantics. The article ``The tractability of subsumption in Frame-Based Description Languages'' by R. Brachman and H. Levesque, addressing the tradeoff between the expressivity of KL-ONE like languages and the computational complexity of reasoning, is usually regarded as the origin for the research on Description Logics.

The research in the area of Description Logics has thereafter begun under the label terminological systems to emphasize the fact that classes and relationships were used to establish the basic terminology adopted in the modeled domain. Later, the emphasis was on the set of concept forming constructs admitted in the language, giving rise to the name concept languages. In the past years, after the attention has been further moved towards the properties of the underlying logical systems, the term Description Logics has become popular.

Since 1989 a dedicated workshop has been held on the subject, initially every two years and from 1994 every year. The Workshops for 1998 and 1999 have already been scheduled. Moreover, sessions on Description Logics have been steadily included in the programs of IJCAI, AAAI, ECAI and KR; Technical articles on Description Logics have appeared in various high-quality journals.

The research on Description Logics has covered theoretical aspects, implementation of knowledge representation systems (modern frame-based systems) and the realization of applications by means of such systems in several areas. This kind of development has been recognized by the Artificial Intelligence community as one example to follow, as far as the methodology of research development is concerned. The key element has been the a very close interaction between theory and practice. On the one hand, there are various implemented systems based on description logics, which offer a palette of description formalisms with differing expressive power, and which are employed in various application domains (such as natural language processing, configuration of technical systems, databases). On the other hand, the formal and computational properties (like decidability, complexity) of various description formalisms are investigated in detail. These investigations are usually motivated by the use of certain constructors in systems or the need for these constructors in specific applications and the results have influenced the design of new systems.

The Description Logics research community currently consists of at least 100 active researchers, the largest groups being in Germany, Italy, and the USA. In addition, other communities are now becoming interested in Description Logics, most notably the Data Base one.

After more than decade of research there is a substantial body of work and well-established material to provide a systematic introduction to the subject. However, a comprehensive introduction to Description Logics is still missing, and neither survey papers available in the literature, nor Workshop Proceedings are adequate for this purpose.