Modelled concept learning for database knowledge discovery

This paper details modelled concept formation phenomena, developed model components and preliminary results of a post-graduate research programme at Griffith University School of Computing and Information Technology. The concept learning model is original and general, and one potential application is as a new and powerful mineral exploration tool. To date, the model has been trialled on an Earth resource database of coal data. This paper extends the understanding of concept learning through a new model, concept learning physics, and applied mathematical formalism. A basis for multivariate knowledge discovery, further research and development is established. Its application adds value to databases through newly discovered theories, new digital knowledge products and intelligent access. Discovered theories improve decision-making with unprecedented hypothesis testing, prediction, logical deductions and explanations.<>