Margin-Based Feature Selection in Incomplete Data

摘要

This study considers the problem of feature selection in in complete data. The intuitive approach is to first impute the missing values, and then apply a standard feature selection method to select relevant features. In this study, we show how to perform feature selection directly, without imputing missing values. We define the objective function of the un certainty margin based feature selection method to maxim ize each instance's uncertainty margin in its own relevant subspace. In optimization, we take into account the uncer tainty of each instance due to the missing values. The exper imental results on synthetic and 6 benchmark data sets with few missing values (less than 25%) provide evidence that our method can select the same accurate features as the al ternative methods which apply an imputation method first. However, when there is a large fraction of missing values (more than 25%) in data, our feature selection method out performs the alternatives, which impute missing values first.