DocumentCode :
2453434
Title :
A Comparative Study of Ensemble Feature Selection Techniques for Software Defect Prediction
Author :
Wang, Huanjing ; Khoshgoftaar, Taghi M. ; Napolitano, Amri
fYear :
2010
fDate :
12-14 Dec. 2010
Firstpage :
135
Lastpage :
140
Abstract :
Feature selection has become the essential step in many data mining applications. Using a single feature subset selection method may generate local optima. Ensembles of feature selection methods attempt to combine multiple feature selection methods instead of using a single one. We present a comprehensive empirical study examining 17 different ensembles of feature ranking techniques (rankers) including six commonly-used feature ranking techniques, the signal-to-noise filter technique, and 11 threshold-based feature ranking techniques. This study utilized 16 real-world software measurement data sets of different sizes and built 13,600 classification models. Experimental results indicate that ensembles of very few rankers are very effective and even better than ensembles of many or all rankers.
Keywords :
data mining; pattern classification; software metrics; software reliability; classification models; comparative study; data mining applications; ensemble feature selection techniques; local optima; multiple feature selection methods; real-world software measurement data sets; signal-to-noise filter technique; single feature subset selection method; software defect prediction; threshold-based feature ranking techniques; Analysis of variance; Data models; Frequency modulation; Measurement; Radio frequency; Software; Training data; defect prediction; ensembles; feature ranking;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Applications (ICMLA), 2010 Ninth International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4244-9211-4
Type :
conf
DOI :
10.1109/ICMLA.2010.27
Filename :
5708824
Link To Document :
بازگشت