First of all, we present how the relationship between problem solving methods and domain models is tackled in different approaches. We concentrate on how they cope with this issue in the knowledge engineering process. Secondly, we introduce several properties which can be used to analyse, characterise and define the notion of role. We evaluate and compare the works exposed previously following these dimensions. This analysis suggests some developments to better exploit the relationship between reasoning and domain knowledge.

In the present paper, we argue that Terminological Knowledge Bases (TKB) are all the more useful for addressing various needs as they do not fulfill formal criteria. Moreover, they intend to clarify the terminology of a given domain by illustrating term uses in various contexts. Thus we designed a TKB structure including 3 linked features: terms, concepts and texts, that present the peculiar use of each term in the domain. Note that concepts are represented into frames whose non-formal description is standardized. Associated with this structure, we defined modeling criteria at the conceptual level. Finaly, we discuss the situation of TKB with regard to ontologies, and the use of TKB for the development of AI systems.

Formal ontology provides axiomatizations of domain independent principles which, among other applications, can be used to identify modeling errors within a knowledge base. The Ontoclean methodology is probably the best-known illustration of this strategy, but its cost in terms of manual work is often considered dissuasive. This article investigates the applicability of such debugging strategies to Description Logics knowledge bases, showing that even a partial and shallow analysis rapidly performed with a top-level ontology can reveal the presence of violations of common sense, and that the bottleneck, if there is one, may instead reside in the resolution of the resulting inconsistency or incoherence.