The importance of negative associations and the discovery of association rule pairs. (29th September 2008)

• Record Type: Journal Article
• Authors: Sim, Alex Tze Hiang; Indrawan, Maria; Srinivasan, Bala
• Is Part Of: International journal of business intelligence and data mining. Volume 3:Number 2(2008)
• Page Start: 158
• Page End: 176
• Publication Date: 2008-09-29
• Subjects: negative association rules -- rule pairs -- implicational support -- association rule mining -- decision making; 006.312
• Languages: English
• ISSNs: 1743-8187

Abstract

Association Rule Mining (ARM) technique is one of the popular data mining techniques used to discover knowledge from a database. In this paper, we highlight the subsequent drawbacks to association rule mining without considering the absence of items. We propose a novel approach to mine both positive and negative association rules as rule pairs. Our approach ensures that impacts of negative associations are considered so that the drawbacks identified can be avoided. Our initial experiment results show that mining pairs of association rules invoking negative associations are small in number and easy to be appreciated for its implications for decision-making.