Performance Analysis of MODENAR and MOEA for Mining Association Rules

Abstract

Evolutionary algorithms have significant impact on association rule mining. These algorithms can be single objective or multi-objective, depending on the intended result. Both Multi-Objective Evolutionary Algorithm (MOEA) and Multi-Objective Differential Evolution Algorithm for Mining Numeric Association Rules (MODENAR) are multi-objective algorithms but uses different objectives. MOEA is an algorithm that mines association rules from a dataset that contains any type of attributes. Contrarily, MODENAR was proposed for mining numeric association rules and this algorithm was not compared to any other algorithms by the authors. To our apprehension, no such work has been done till now that compares MODENAR with other association rule mining algorithms. So, this paper will compare MODENAR with another association rule mining algorithm, named MOEA and demonstrate the results by applying above mentioned algorithms on different datasets.