Alexandre I. Direne
This thesis describes how high-level knowledge about visual images can be represented and further interpreted through system-active and system-passive tutorial interactions. The ideas lend themselves to the design and implementation of a category of Intelligent Tutoring Systems (ITSs) aimed at the teaching of abnormalities in highly visual domains like medical diagnostic imaging. Most past work in visual concept tutoring has concentrated on the theoretical principles of how humans acquire expertise in visual recognition. The few implementations there have been are domain-specific.
This paper is not available online