Two Improvements to C4.5 Algorithm
Weixiang Sun · Journal of Jiangsu Polytechnic University · 2008
As an important decision tree algorithm,C4.5 still has two disadvantages.One is that it is very time-consuming to find the optimal threshold of continuous attribute.The other is that C4.5 has no ability of incremental learning.Based on the boundary point theorem given by Fayyad,an improvement method of selecting the optimal threshold is proposed to overcome the first disadvantage.Also,after the modification of tree structure of C4.5,an incremental C4.5 algorithm is put forward to solve the second problem.