基于双向有序互信息的单调分类决策树算法

南京大学学报(自然科学版) ›› 2013, Vol. 49 ›› Issue (5): 628–636.

基于双向有序互信息的单调分类决策树算法

许行1，梁吉业1,2，王宝丽1

出版日期:2014-02-08 发布日期:2014-02-08
作者简介:(计算智能与中文信息处理教育部重点实验室，山西大学计算机与信息技术学院，太原，030006; 2. 太原师范学院计算机系，太原，030012 )
基金资助:
973计划前期研究专项（2011CB311805），山西省科技基础条件平台建设项目（2012091002-0101）

Bi-direction ank mutual information based decision trees for monotonic classification

Xu Hang1, Liang Ji-Ye1,2, Wang Bao-Li1

Online:2014-02-08 Published:2014-02-08
About author:(1. Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, School of Computer and Information Technology, Shanxi University, Taiyuan, 030006, China; 2. Department of Computer, Taiyuan Normal University, Taiyuan, 030012, China)

摘要/Abstract

摘要： 决策树是一种智能进行实例分类的数据挖掘方法，已被广泛应用于机器学习、数据挖掘、智能控制等人工智能领域。单调决策树可以解决属性具有单调序关系的分类问题，近年来引起了国内外研究者的广泛关注。Hu提出了基于优势关系的有序信息熵的概念，并将其成功地运用于有序决策树的构造算法中，得到了较好的效果。在Hu的算法的基础上，利用双向的有序互信息生成不同的决策树，再集成其分类规则得到最后的决策结果，实验数据表明，相对于单向的有序分类树，此算法可以提高分类准确率，缩短分类规则的长度。

Abstract: The decision tree is an intelligent data mining method for instance classification. It has been widely used in the artificial intelligence field of machine learning, data mining, intelligent control, and so on. Decision Trees for Monotonic Classification can resolve classification problems that its attributes has a monotonous rank relationship. In recent years, many researchers have paid their attention to this kind of problems. Hu et al. designed the concept of rank information entropy based on dominance relations, and then they successfully applied it in the construction algorithm of rank decision tree and got better results. In this paper we generated different decision trees for monotonic classification based on the Bi-direction rank mutual information, then integrated the classification rules to get the final decision-making results. Compared with the unidirectional rank decision tree, the proposed algorithm can improve the classification accuracy and shorten the length of the classification rules

许行1，梁吉业1,2，王宝丽1. 基于双向有序互信息的单调分类决策树算法[J]. 南京大学学报(自然科学版), 2013, 49(5): 628–636.

Xu Hang1, Liang Ji-Ye1,2, Wang Bao-Li1. Bi-direction ank mutual information based decision trees for monotonic classification[J]. Journal of Nanjing University(Natural Sciences), 2013, 49(5): 628–636.

参考文献

[1] Quinlan J. R. Induction of decision trees. Machine learning, 1986, 1(1): 81~106.

[2] Quinlan J R. C4. 5: Programs for machine learning. Morgan Kaufmann Publisher Inc, 1993, 302.

[3] Ben-David A, Sterling L, Pao Y H. Learning and classification of monotonic ordinal concepts. Computational intelligence, 1989, 5: 45~49.

[4] Ben-David A. Monotonicity maintenance in information theoretic machine learning algorithms. Machine Learning, 1995, 19: 29~43.

[5] Ben-David A, Sterling L, Tran T. Adding monotonicity to learning algorithms may impair their accuracy ． Expert Systems with Applications, 2009, 36: 6627~6634.

[6] Kotlowski W, Slowihski R. Rule learning with monotonicity constraints. Proceeding of the 26^th International Conference on Machine Learning, Montreal, Canada, 2009: 537~544.

[7] Liang J Y, Qian Y H. Information granules and entropy theory in information systems. Science in China., 2008, 38(2): 2048~2065. （梁吉业 , 钱宇华 . 信息系统中的信息粒与熵理论 . 中国科学 , 2008, 38(2): 2048~2065 ） .

[8] Wang B L, Liang J Y. Knowledge distance and rough entropy in information systems. Computer Science, 2007(3): 151~154. （王宝丽 , 梁吉业 . 信息系统中的知识距离与知识粗糙熵 . 计算机科学 , 2007(3) : 151~154 ） .

[9] Liang J Y, Li D Y. The uncertainty and knowledge acquisition in information systems. Beijing: Science Publishing, 2005, 118. （梁吉业 , 李德玉 . 信息系统中的不确定性与知识获取 . 北京：科学出版社 , 2005, 118. ）

[10] Liang J Y, Shi Z Z. The information entropy, rough entropy and knowledge granulation in rough set theory. International Journal of Uncertainty, Fuzziness and Knowledge-based Systems, 2004, 12(1): 37~46.

[1 1 ] Hu Q H,Guo M Z, Yu D R, et al. Information entropy for ordinal classification. Science in China Series F: Information Sciences. 2010, 53(6):1188~1200.

[12] Hu Q H, Pan W W, Zhang L, et al. Feature selection for monotonic classification. IEEE Transactions on Fuzzy Systems, 2012, 20(1): 69~81.

[13] Hu Q H, Che X J, Zhang L, et al. Rank entropy based decision trees for monotonic classification. IEEE Transactions on Knowledge and Data Engineering, 2012, 24(11): 2052~2064.

[14] Che X J. Ordinal decision tree based fault level detection. Dissertation for the Master Degree in Engineering. Harbin: Harbin Institute of Technology, 2011. （车勋建 . 基于有序决策树的故障程度诊断研究 . 硕士学位论文 . 哈尔滨：哈尔滨工业大学 , 2011. ） .

[15] Greco S, Matarazzo B, Slowinski R. Rough approximation by dominance relations. Intenational Journal of Intelligent Systems 2002, 17:153~171.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed