An improved TANC classification algorithm based on C4.5

doi:10.1109/CCDC.2014.6853067

	An improved TANC classification algorithm based on C4.5
	Zhao, Xiao-Qiang 1; Yang, Jia-Min 1,2
	2014
会议名称	26th Chinese Control and Decision Conference, CCDC 2014
会议录名称	26th Chinese Control and Decision Conference, CCDC 2014
页码	4992-4996
会议日期	May 31, 2014 - June 2, 2014
会议地点	Changsha, China
出版者	IEEE Computer Society
摘要	Tree Augmented Naive Bayes Classification (TANC) is not very well to deal with continuous data and it ignores partial data in the absence of data attribute value and this can reduce the result accuracy. To resolve this problem, an improved algorithm based on C4.5 is proposed in this paper. The proposed algorithm firstly modifies the available training data according to the predictions of C4.5, then continuous data is discretized by dividing many finite intervals of attributes, this modified training data is used to train TANC. In this way it can improve the classification accuracy of the TANC. The experimental results show that the improved algorithm is superior to TANC in terms of classification accuracy. © 2014 IEEE.
关键词	Classification (of information) Classifiers Forestry Learning systems Machine learning C4.5 algorithm Classification accuracy Classification algorithm Continuous data Data attributes Finite intervals Training data Tree augmented Naive Bayes
DOI	10.1109/CCDC.2014.6853067
收录类别	EI
语种	中文
EI入藏号	20143218039919
EI主题词	Trees (mathematics)
来源库	Compendex
分类代码	716.1 Information Theory and Signal Processing - 802.1 Chemical Plants and Equipment - 921.4 Combinatorial Mathematics, Includes Graph Theory, Set Theory
引用统计	无
文献类型	会议论文
条目标识符	https://ir.lut.edu.cn/handle/2XXMBERH/117774
专题	兰州理工大学
作者单位	1.College of Electrical and Information Engineering, Lanzhou Univ. of Tech., Lanzhou 730050, China; 2.Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou 730050, China
推荐引用方式 GB/T 7714	Zhao, Xiao-Qiang,Yang, Jia-Min. An improved TANC classification algorithm based on C4.5[C]:IEEE Computer Society,2014:4992-4996.