From educational data mining model to the automated knowledge based system construction

Abstract

A knowledge based system (KBS) has its advantage over conventional database systems in that it has the inference ability to deduce implicit knowledge from the explicitly stored information. KBS is however known to be labor intensive in its construction in the knowledge acquisition and elicitation phase. Researchers have tried to overcome this hindrance for more than four decades. Automatic creation of a knowledge base (KB) content is still a research topic of interest. In this paper, we propose the design of a framework that not only automatically creates a KB, but also constructs the inference and reasoning engine of the KBS. The KB content is elicited and transferred from the data mining model, whereas the engine (or shell) of the KBS is created from the decision rules. We demonstrate a case study in student loan payment decision using the visualized tools KNIME and WIN-PROLOG to generate a data mining model and a KBS shell, respectively.