An efficient visual classification based approach to Decision Tree construction

Decision Tree (DT) analysis has emerged over the decades as an effective tool in classification or prediction. Since the publication of the first comprehensive and authoritative book on decision tree analysis by Howard Raiffa in 1968, its applications to a variety of problems from numerous disciplines have grown enormously. However, most of the methods for DT construction have some pitfalls including binary split points of numeric attributes instead of arbitrary splitting, involvement of users with prior domain knowledge to construct DT and finally the absence of training data visualization which this paper aims to remove. Besides we have proposed how the constructed DT can be applied for the adaptive interview process.