Cluster and Classify: A Conceptual Approach

From the platform of a data set that was gathered in a complex problem solving task, this paper presents preliminary results of the evaluation of the feasibility of establishing conceptual clustering schemes within the structure of an Object Oriented Programming System, namely KEE (Knowledge Engineering Environment). It is shown that both adequate data description and representation are crucial in order to successfully develop the method. The conclusion is that it is feasible to generate description primitives from the data. However, it is not quite clear yet how to evaluate the quality of potential solutions of an inductive process.