Hybridization of ABC and PSO for Optimal Rule Extraction from Knowledge Discovery Database

Abstract

Knowledge discovery in database (KDD) has provided a large interest in statistics, machine learning, and artificial intelligence (AI). It is a challenging task for mining the comprehensive and informative knowledge in such complex data by using the existing methods. The challenges come from many aspects, for instance, the traditional methods usually discover homogeneous features from a single source of data while it is not effective to mine for patterns combining components from multiple data sources. It is often very costly and sometimes impossible to join multiple data sources into a single data set for pattern mining. In order to extract knowledge from different datasets, we will propose a hybrid mining technique. The knowledge extraction can be done by association rule mining with the combination of Artificial Bee Colony optimization algorithm (ABC) and Particle swarm optimization (PSO). The main aim of this hybridization is to extract the optimal rules from the association rules for further classification. The accuracy will be checked in terms of optimal rule obtained from the hybridization. The best position for moving the particle will be updated by using ABC algorithm.