This article provides a brief introduction to the concepts of feature selection. In the other subsections of this chapter we address several topics related to feature selection.
There are three main types of features selection: (i) feature scoring, (ii) feature ranking and (iii) feature subset selection. Feature scoring is the most general method and can be converted in the latter two, while feature ranking can only be turned into feature subset selection methods.
Any of these three types of feature selection can be converted to an ensemble feature selection method. Currently Java-ML only provides an ensemble of feature rankers, but in the future the other two will be supported as well.