Browse > Article
http://dx.doi.org/10.5351/KJAS.2004.17.2.347

A Study on Selection of Split Variable in Constructing Classification Tree  

정성석 (전북대학교 통계정보과학과)
김순영 (전북대학교 통계정보과학)
임한필 (전북대학교 통계정보과학과)
Publication Information
The Korean Journal of Applied Statistics / v.17, no.2, 2004 , pp. 347-357 More about this Journal
Abstract
It is very important to select a split variable in constructing the classification tree. The efficiency of a classification tree algorithm can be evaluated by the variable selection bias and the variable selection power. The C4.5 has largely biased variable selection due to the influence of many distinct values in variable selection and the QUEST has low variable selection power when a continuous predictor variable doesn't deviate from normal distribution. In this thesis, we propose the SRT algorithm which overcomes the drawback of the C4.5 and the QUEST. Simulations were performed to compare the SRT with the C4.5 and the QUEST. As a result, the SRT is characterized with low biased variable selection and robust variable selection power.
Keywords
Classification tree; Grouping; Peizer ＆ Pratt transformation; Variable selection bias; Variable selection power;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Improved use of continuous attribute in C4.5 /
[ Quinlan, J. R. ] / Journal of Artificial Intelligence Research
2 An Exploratory technique for investigating large quantities of categorical data /
[ Kass, G. V. ] / Applied Statistics   DOI   ScienceOn
3 Multiway Split Classification Trees /
[ Kim, H. ] / Ph.D. Thesis, University of Wisconsin
4 Classification trees with unbiased multiway splits /
[ Kim, H.;Loh, W. Y. ] / Journal of the American Statistical Association   DOI   ScienceOn
5 Split selection method for classification trees /
[ Loh, W. Y.;Shih, Y. S.; ] / Statistica Sinica
6 Tree-structured classification via generalized discriminant analysis (with discussion) /
[ Loh, W. Y.;Vanichsetakul, N. ] / Journal of the American Statistical Association   DOI   ScienceOn
7 /
[ Quinlan, J. R. ] / C4.5 : Programs for Machine Learning
8 /
[ Witten, I. H.;Frank, E. ] / Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations
9 A study on bias problems in constructing classification trees /
[ 이윤모 ] / 서울대학교 박사학위논문
10 /
[ Breiman, L.;Friedman, J. H.;Olshen, R. A.;Stone, C. J. ] / Classification and Regression Trees