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 Rènyi, 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.