REVIEW ON METAHEURISTIC-BASED DECISION TREE INDUCTION

Decision tree is one of the well-known machine learning algorithms which remain popular due its simple implementation and ease of understanding. Existing techniques used in inducing decision tree have shown that they suffer data overfitting which leads to producing small size of decision tree model. To address the drawback and improve the performance of the algorithm, hybridization of metaheuristic algorithm and decision tree are done by many researchers. In this article, we undertake a review on the hybrid done with swarm intelligence on decision tree based on ant-miner and other different modifications. Different application domains executed using metaheuristic-based decision tree are described. Finally, we address some challenges and future research directions.