Bipartition techniques for quantitative attributes in association rule mining

In this paper we propose a systematic approach to mine quantitative association rules-association rules which contain quantitative attributes-using commercial mining tools. To achieve this goal, we first propose an overall working framework that consists of two steps: (1) a pre-processing step which converts quantitative attributes into binary attributes and (2) a post-processing step which reconverts binary association rules into quantitative association rules. We then formally redefine the previous mean-based and median-based bipartition techniques. These previous bipartition techniques, however, have the problem of not considering distribution characteristics of attribute values. To solve this problem, we propose an intuitive bipartition technique, named standard deviation minimization, which divides a quantitative attribute into two partitions to minimize their standard deviations. Through extensive experiments, we argue that our framework works correctly, and we show that our standard deviation minimization is superior to other bipartition techniques.