Consistency Based Feature Selection

摘要: Feature selection is an effective technique in dealing with dimensionality reduction for classification task, a main component of data mining. It searches "optimal" subset features. The search strategies under consideration are one the three: complete, heuristic, and probabilistic. Existing algorithms adopt various measures to evaluate goodness feature subsets. This work focuses on measure called consistency. We study its properties comparison other major different ways using this conduct empirical examine pros cons these methods Through extensive exercise, we aim provide comprehensive view relations guideline use facing new application.