An algorithm of decision tree learning based on attribute contribution
Xuegang Hu · Journal of Hefei University of Technology · 2009
An algorithm of decision tree learning based on attribute contribution,which is called the ICD algorithm,is proposed.The algorithm constructs the model of attribute contribution functions based on certain classic information distributing by the information entropy theory.It eliminates the bias of selecting the attribute with many values by the best attribute choice.The comparison among the algorithms ICD,ID3 and C4.5 is made by experiments,and the results indicate that the ICD algorithm has better performance.