SONAR: System for Optimized Numeric Association Rules

Abstract

A system called System for Optimized Numeric Association Rules (SONAR) for mining optimized association rules from databases with numeric data as well as Boolean data is introduced. SONAR focuses on computing an optimized support (confidence) range that maximizes the support on the condition that the confidence ratio is at or above a given threshold. SONAR also generates a two-dimensional version of association rules.