Instance Reduction Using Multi-Objective Chaotic Particle Swarm Optimization Algorithm

Today, it is important to reduce the original huge data set to a manageable volume. Also, unbalanced data distribution between different classes is a serious challenge in data mining. In the proposed method, the instance reduction problem is considered as a multi-objective problem, which can perform well by considering the two contradict criteria, classification accuracy and reduction rate of instances. The multi-objective problem is solved using the chaotic particle swarm optimization algorithm. The distance-based decision classifier has the task of distinguishing the maintenance or deletion of test instances. Creating and maintaining balances for different types of data distribution is the main goal of the proposed method. The results of the experiments have been compared with the state-of-the-art methods, which show superiority of the proposed method in terms of classification accuracy and reduction percentage.The results of the experiments have been compared with the state-of-the-art methods, which show superiority of the proposed method in terms of classification accuracy and reduction percentage.