Mining confident minimal rules with fixed-consequents

Abstract

Association rule mining (ARM) finds all the association rules in data that match some measures of interest such as support and confidence In certain situations where high support is not necessarily of interest, fixed-consequent association-rule mining for confident rules might be favored over traditional ARM. The need for fixed consequent ARM is becoming more evident in a number of applications such as market basket research (MBR) or precision agriculture. Highly confident rules are desired in all situations; however, support thresholds fluctuate with the applications and the data sets under study as we shall show later. In this paper 1 , we propose an approach for mining minimal confident rules in the context of fixedconsequent ARM that relieves the user from the burden of specifying a minimum support threshold. We show that the framework suggested herein is efficient and can be easily expanded by adding new pruning conditions pertaining to specific situations.

Episode Rules Mining Algorithm for Distant Event Prediction

“…Traditional episode rules mining algorithms construct episode rules with a large antecedent (made up of many events) (Pasquier et al, 1999). Discovering rules with a small antecedent was introduced for association rules, called "minimal association rules discovery" (Rahal et al, 2004). Minimal rules are considered as sufficient (as no new knowledge is given by larger ones).…”