A knowledge-based system is used as a front-end to a very large database to increase the relevance of the information being retrieved. The subject domain of the data base is modelled in a semantic network and the queries to the database are expanded according to the semantic model. An experiment has been performed on a bibliographic database, by developing the prototype KNOWIT, a knowledge-based front-end to the information retrieval system ESA-QUEST11The prototype was developed for the European Space Agency under ESA-Contract AO/1-2096/89/HGE-1.. An experimental evaluation shows that the number of relevant bibliographic references retrieved with the knowledge-based front-end is significantly improved, without compromising the precision of the retrieval.

The METAKREK project addresses the important problem of knowledge acquisition, which is a major bottleneck in the development of knowledge-based systems. The project resulted in a methodology for knowledge-base modeling (METAMETH) and a tool kit for knowledge-base prototyping and testing (METATOOL). The methodology emphasize the importance of knowledge modeling a and propose modeling at different levels, while the development of a graphical interface for visualization of knowledge is the major achievement of the tool kit. The paper describes the modeling approach applied on a medical application (interpretation of serum protein electrophoresis) and lists the experiences from applying the methodology and the tool kit.