论文信息 - Separability of Split Value Criterion with Weighted Separation Gains

Separability of Split Value Criterion with Weighted Separation Gains

An analysis of the Separability of Split Value criterion in some particular applications has led to conclusions about possible improvements of the criterion. Here, the new formulation of the SSV criterion is presented and examined. The results obtained for 21 different benchmark datasets are presented and discussed in comparison with the most popular decision tree node splitting criteria like information gain and Gini index. Because the new SSV definition introduces a parameter, some empirical analysis of the new parameter is presented. The new criterion turned out to be very successful in decision tree induction processes.

Krzysztof Grabczewski

[1] Wodzisaw Duch,et al. THE SEPARABILITY OF SPLIT VALUE CRITERION , 2000 .

[2] Norbert Jankowski,et al. Versatile and Efficient Meta-Learning Architecture: Knowledge Representation and Management in Computational Intelligence , 2007, 2007 IEEE Symposium on Computational Intelligence and Data Mining.

[3] Wray L. Buntine,et al. A Further Comparison of Splitting Rules for Decision-Tree Induction , 1992, Machine Learning.

[4] Krzysztof Grabczewski,et al. Efficient and friendly environment for computational intelligence , 2010 .

[5] K. Gr,et al. A general purpose separability criterion for classification systems , 1999 .

[6] Leo Breiman,et al. Classification and Regression Trees , 1984 .

[7] John Mingers,et al. An empirical comparison of selection measures for decision-tree induction , 2004, Machine Learning.

[8] Aiko M. Hormann,et al. Programs for Machine Learning. Part I , 1962, Inf. Control..