Colon Cancer Stage Classification Using Decision Trees

Abstract

Decision tree methodology is one of the most commonly used data mining techniques and is used for both classification and regression problems. The objective of this research is to perform a comparative study on stage classification of colon cancer data based on multiple covariates using various decision tree algorithms and to identify the best classifier for stage classification of colon cancer data. Decision tree algorithms are non-parametric and can efficiently deal with large datasets. We performed the analysis by splitting the data under study into training and testing datasets. This paper utilizes frequently used decision tree algorithms including CART, QUEST, and CHAID and identify on the appropriate tree size needed to achieve the optimal final model.