 The CABC algorithm is a novel approach to feature selection which reduces the dimensionality of a dataset while preserving the most important features. It does so by identifying the most relevant features using a combination of Lorentz curve analysis and item categorization. The algorithm was tested on three different datasets, each containing hundreds of thousands of variables, and was able to reduce the size of the dataset without sacrificing accuracy. Additionally, the algorithm has the ability to stop when it reaches a certain level of accuracy, making it more efficient than other methods. This article was authored by John Lodge and Alfred Olch.