Mining Contrast Inequalities in Numeric Dataset
Finding relational expressions which exist frequently in one class of data while not in the other class of data is an interesting work. In this paper, a relational expression of this kind is defined as a contrast inequality. Gene Ex pression Programming (GEP) is powerful to discover relations from data and express them in mathematical level. Hence, it is desirable to apply GEP to such mining task. The main contributions of this paper include: (1) introducing the concept of contrast inequality mining, (2) designing a two-genome chromosome structure to guarantee that each individual in GEP is a valid inequality, (3) pro posing a new genetic mutation to improve the efficiency of evolving contrast inequalities, (4) presenting a GEP-based method to discover contrast inequalities, (5) giving an extensive performance study on realworld datasets. The experi mental results show that the proposed methods are effective. Contrast inequali ties with high discriminative power are discovered from the real-world datasets. Some potential works on contrast inequality mining are discussed.
Data Mining Contrast Mining Gene Expression Programming
Lei Duan Jie Zuo Tianqing Zhang Jing Peng Jie Gong
School of Computer Science, Sichuan University, Chengdu 610065, China Sci. & Tech. Department, Chengdu Municipal Public Security Bureau, Chengdu 610017, China
国际会议
11th International Conference,WAIM 2010(第十一届网络时代管理国际会议)
九寨沟
英文
194-205
2010-07-14（万方平台首次上网日期，不代表论文的发表时间）