MMDT: Multi-Objective Memetic Rule Learning from Decision Tree

In this article, a Multi-Objective Memetic Algorithm (MA) for rule learning is proposed. Prediction accuracy and interpretation are two measures that conflict with each other. In this approach, we consider accuracy and interpretation of rules sets. Additionally, individual classifiers face other problems such as huge sizes, high dimensionality and imbalance classes’ distribution data sets. This article proposed a way to handle imbalance classes’ distribution. We introduce Multi-Objective Memetic Rule Learning from Decision Tree (MMDT). This approach partially solves the problem of class imbalance. Moreover, a MA is proposed for refining rule extracted by decision tree. In this algorithm, a Particle Swarm Optimization (PSO) is used in MA. In refinement step, the aim is to increase the accuracy and ability to interpret. MMDT has been compared with PART, C4.5 and DTGA on numbers of data sets from UCI based on accuracy and interpretation measures. Results show MMDT offers improvement in many cases.