The Application of an Improved C4.5 Decision Tree

Abstract

Decision tree classification method is an effective instance-based learning and data-mining method. In this paper, several decision tree classification algorithms were analyzed, including ID3 and C4.5 algorithm, followed by some improved algorithms. Taylor series and Maclaurin's series are applied to simplify the information gain ratio formula in which way the time efficiency of calculation can be improved. The improved C4.5 algorithm can be applied in the student employment recommendation system. Through this improved C4.5 decision tree algorithm, graduates could have a good guide while seeking for a job and teachers could also help their students in decision-making and job recommendation.