Fuzzy-Rough Feature Selection

In this chapter, the theoretical developments behind this new feature selection method are presented together with a proof of generalization. This novel approach uses fuzzy-rough sets to handle many of the problems facing feature selectors outlined previously. A complexity analysis of the main selection algorithm is given. The operation of the approach and its bene ts are shown through the use of two simple examples. To evaluate this new fuzzy-rough measure of feature signi cance, comparative investigations are carried out with the current leading significance measures.

R. Jensen and Q. Shen. Computational Intelligence and Feature Selection: Rough and Fuzzy Approaches. IEEE Press/Wiley & Sons.