From knowledge modelling to expertise elicitation in cytology

This paper deals with the design of a multi-agent knowledge acquisition environment dedicated to cytology. This environment is intended to remain compatible with a robust theory of knowledge and the distributed artificial Intelligence concepts. The design, based on three cooperating modes (elicitation, presentation and resolution), is first of all introduced. Two elicitation forms are then described, which fit the cytological expertise. Finally, the user interface is presented before discussing the main points of the approach.