ImbTreeEntropy: An R package for building entropy-based classification trees on imbalanced datasets

In this paper, we propose a novel R package, named ImbTreeEntropy, for building binary and multiclass decision trees using generalized entropy functions, such as Renyi, Tsallis, Sharma-Mittal, Sharma-Taneja and Kapur, to measure the impurity of a node. These are important extensions of the existing algorithms that usually employ Shannon entropy and the concept of information gain. Additionally, ImbTreeEntropy is able to handle imbalanced data, which is a challenging issue in many practical applications. The package supports cost-sensitive learning by defining a misclassification cost matrix and weighted sensitive learning. It accepts all types of attributes, including continuous, ordered and nominal attributes. The package and its code are made freely available. (C) 2021 The Authors. Published by Elsevier B.V.