Exante: A Preprocessing Method for Frequent-Pattern Mining

Francesco Bonchi, Fosca Giannotti, Alessio Mazzanti, Dino Pedreschi

Abstract: Our main research objective is to define a data mining query language, supported by a system that can optimize constraint-based data mining queries. We have invented ExAnte, a simple yet effective preprocessing technique for frequent-pattern mining. ExAnte exploits constraints to dramatically reduce the analyzed data to those containing patterns of interest. This data reduction, in turn, induces a strong reduction of the candidate patterns' search space, thus supporting substantial performance improvements in subsequent mining.