Im-IG:a novel feature selection method for imbalanced problems
Guozheng Li · Journal of Shandong University · 2010
An imbalanced data set is an ubiquitous problem in the machine learning field,which attracts much attention from related scientists.The information gain(IG)method is widely used in feature selection,but it is seldom studied in imbalanced problem.Based on the performance discussion of IG on imbalanced data sets,a new method Im-IG was proposed for the imbalanced problem in feature selection.Im-IG increased the weight of minor class in the entropy calculation,in order to select features which were better for minor class.Im-IG focused on improving the classification accuracy of minor class,based on the performance improvement of the whole data set.Experimental results on several imbalanced data sets showed that Im-IG can solve the imbalanced predicament IG met and it was an effective feature selection method for the imbalanced problem.