Data mining by means of generalized patterns

The thesis is mainly focused on the study and the application of pattern discovery algorithms that aggregate database knowledge to discover and exploit valuable correlations, hidden in the analyzed data, at different abstraction levels. The aim of the research effort described in this work is two-fold: the discovery of associations, in the form of generalized patterns, from large data collections and the inference of semantic models, i.e., taxonomies and ontologies, suitable for driving the mining process.