Interactive Exploration of Discovered Knowledge: A Methodology for Interaction, and Usability Studie

The papers in the series are intended for internal use and are distributed by the author. Copies may be ordered from the library of Department of Computer Science. Abstract We introduce a methodology for knowledge discovery in databases (KDD) where one rst discovers large collections of patterns at once, and then performs interactive retrievals from the collection of patterns. The proposed methodology suits very well such KDD formalisms as association and episode rules, where large collections of potentially interesting rules can be found ef-ciently. We present methods that support interactive exploration of large collections of association and episode rules. With these methods the user can exibly specify the focus of interest, and also iteratively reene it. We claim that the proposed methodology is very usable for a class of knowledge discovery problems. We describe an implementation of the explorative part. We discuss some usability aspects and usability testing of KDD systems, and present an actual usability test of the described implementation.