Studi Algoritma Cart dengan Induksi Fuzzy dalam Mengklasifikasikan Data

Abstract

Decision tree methods used for extracting information about dataset in classification problems, although unable to deal with uncertainties embedded within the data associated with human thinking and perception. This paper describes the development of a tree induction algorithm which improves the classification accuracy of decision tree induction. The research is designed by integrating the principles of CART (Classification and Regression Tree) algorithm and the fuzzy set-theoretic concepts, enabling the model to handle uncertain and imprecise data, and in order to soften the sharp decision boundaries which are inherent in crisp decision tree algorithms. CART is a decision tree algorithm with the main feature of Gini Index testing (homogenity index) at each level, leading to the production of trees with pruning process to find the optimal tree. The application of fuzzy logic to CART decision trees can represent classification knowledge more naturally and inline with human thinking, and are more robust when it comes to handling imprecise information. The results of applying fuzzy logic to CART decision trees are presented in this paper. These have been obtained from sets of real data, and show that the new fuzzy inference algorithm improves the accuracy over crisp CART trees.