2010 | OriginalPaper | Buchkapitel
EWGen: Automatic Generation of Item Weights for Weighted Association Rule Mining
verfasst von : Russel Pears, Yun Sing Koh, Gillian Dobbie
Erschienen in: Advanced Data Mining and Applications
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
Association Rule Mining is an important data mining technique that has been widely used as an automatic rule generation method. While having outstanding success in many different application domains, it also has the potential to generate a vast number of rules, many of which are of little interest to the user. Weighted Association Rule Mining (WARM) overcomes this problem by assigning weights to items thus enabling interesting rules to be ranked ahead of less interesting ones and making it easier for the user to determine which rules are the most useful. Past research on WARM assumes that users have the necessary knowledge to supply item weights. In this research we relax this assumption by deriving item weights based on interactions between items. Our experimentation shows that the rule bases produced by our scheme produces more compact rule bases with a higher information content than standard rule generation methods.