Association Rule Mining Using Discrete Jaya Algorithm

Abstract

Data Mining (DM) is defined as the non-trivial extraction of implicit, previously unknown and potentially useful information from data. Association Rule Mining (ARM) is one of the most important and well studied task in DM for discovering the most interesting patterns from the data stored in databases. Existing approaches suffer from problems of redundant computation, high time complexity, and large storage space. Motivated by the success of population-based metaheuristic in spite of large transactional databases, we proposed an ARM approach based on a Discrete JAYA population-based metaheuristic. This paper formulates the problem of ARM based on Discrete Jaya Algorithm (DJaya-ARM). The effectiveness of the proposed approach is tested on a real database of a population with or without Diabetes Mellitus and compared to FP-Growth Algorithm.