Interval-valued fuzzy-rough feature selection in datasets with missing values

One of the many successful applications of rough set theory has been to the area of feature selection. The rough set principle of using only the supplied data and no other information has many benefits, where most other methods require supplementary knowledge. Fuzzy-rough set theory has recently been proposed as an extension of this, in order to better handle the uncertainty present in real data. However, following this approach, there has been no investigation (theoretical or otherwise) into how to deal with missing values effectively, another problem encountered when using real world data. This paper proposes an extension of the fuzzy-rough feature selection methodology, based on interval-valued fuzzy sets, as a means to counter this problem via the representation of missing values in an intuitive way.