An empirical study on attribute selection measures in decision tree learning

作者：Jiang Liangxiao; Li Chaoqun^*

来源：Journal of Computational Information Systems, 2010, 6(1): 105-112.

• Attribute selection
• Classification
• Classification decision
• Critical problems
• Data sets
• Decision tree learning
• Empirical studies
• In-buildings
• Ranking performance
• Real world data

In many real-world data mining applications, decision trees are widely used for classification and ranking. A critical problem in building decision trees is the attribute selection measure problem. In this paper, we survey all kinds of measures for selecting the best split and then provide an empirical study on the classification and ranking performance of the decision trees using different attribute selection measures. The experimental results based on a large number of UCI datasets verify our conclusions drawn in this paper.

• 出版日期2010