Learn++.MF: A random subspace approach for the missing feature problem

摘要: We introduce Learn^+^+.MF, an ensemble-of-classifiers based algorithm that employs random subspace selection to address the missing feature problem in supervised classification. Unlike most established approaches, Learn^+^+.MF does not replace values with estimated ones, and hence need specific assumptions on underlying data distribution. Instead, it trains ensemble of classifiers, each a subset available features. Instances are classified by majority voting those classifiers whose training did include show can accommodate substantial amount data, only gradual decline performance as increases. also analyze effect cardinality subsets, size performance. Finally, we discuss conditions under which proposed approach is effective.