In the machine learning literature discretization methods have been categorized into two groups: supervised and unsupervised discretization. The first of the unsupervised discretization method is equal interval width discretization, where the range of observed values is divided into k internals of equal length as follows: