Feature transformation methods in data mining

被引：71
作者：
Kusiak, A [1 ]
机构：
[1] Univ Iowa, Dept Mech & Ind Engn, Intelligent Syst Lab, Iowa City, IA 52242 USA
来源：
IEEE TRANSACTIONS ON ELECTRONICS PACKAGING MANUFACTURING | 2001年 / 24卷 / 03期
关键词：
classification; data mining; decision making; feature bundling; feature transformation method; knowledge discovery; transformed data set;
D O I：
10.1109/6104.956807
中图分类号：
T [工业技术];
学科分类号：
08 ;
摘要：
The quality of knowledge extracted from a data set can be enhanced by its transformation. Discretization and filling missing data are the most common forms of data transformation. A new transformation method named feature bundling is introduced. A feature bundle involves a set of features in its pure or transformed form. The computational results reported in this paper show that the classification accuracy of decision rules generated from data sets with feature bundles is enhanced. The proposed concept of feature bundling is applied to a data set from semiconductor industry.
引用
收藏
页码：214 / 221
页数：8
相关论文
共 27 条
[1]   TOLERATING NOISY, IRRELEVANT AND NOVEL ATTRIBUTES IN INSTANCE-BASED LEARNING ALGORITHMS [J].
AHA, DW .
INTERNATIONAL JOURNAL OF MAN-MACHINE STUDIES, 1992, 36 (02) :267-287
[2]  
AUER P, 1995, P 8 EUR C MACH LEARN
[3]  
BOOKER LB, 1989, PROCEEDINGS OF THE THIRD INTERNATIONAL CONFERENCE ON GENETIC ALGORITHMS, P265
[4]  
Cherkassky V, 1997, IEEE Trans Neural Netw, V8, P1564, DOI 10.1109/TNN.1997.641482
[5]  
Cherkassky V.S., 1998, LEARNING DATA CONCEP, V1st ed.
[6]  
Clark P., 1989, Machine Learning, V3, P261, DOI 10.1023/A:1022641700528
[7]  
Domingos P., 1996, Machine Learning. Proceedings of the Thirteenth International Conference (ICML '96), P105
[8]  
DONNART JY, 1994, COM ADAP SY, P144
[9]  
FRIEDMAN J, 1996, P 13 NAT C ART INT C
[10]  
Goldberg D. E., 1989, GENETIC ALGORITHMS S
123