2010 | OriginalPaper | Buchkapitel
Measures for Comparing Association Rule Sets
verfasst von : Damian Dudek
Erschienen in: Artificial Intelligence and Soft Computing
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Most experimental methods for evaluating algorithms of association rule mining are based solely on quantitative measures such as correlation between minimum support, number of rules or frequent itemsets and data processing time. In this paper we present new measures for comparing association rule sets. We show that observing rule overlapping, support and confidence in two compared rule sets helps evaluate algorithm quality or measure uniformity of source datasets.