Improved Accuracy for C4.5 Decision Tree Algorithm

Decision tree is probably the most widely used approach to represent classifiers. Originally it has been studied in the field of decision theory and statistics. However, it was found to be effective in other disciplines such as data mining, machine learning and pattern recognition. This research deals with the problem of finding the parameter settings of decision tree algorithm in order to reach more accuracy for a given domain. The proposed approach Improved C4.5 (IC4.5) is a supervised learning model based on C4.5 algorithm to construct a decision tree. The modification on C4.5 algorithm includes using improved gain instead of gain ratio measure to choose the best attribute and increase the accuracy of decision tree. Our algorithm has been experimented some data sets from UCI repository. The results obtained from experiments show that the accuracy of IC4.5 is greater than C4.5 in increasing the accuracy of decision tree.

کلیدواژه ها:

decision tree ، data mining ، c4.5 algorithm ، accuracy ، classifiers