Fuzzy-rough nearest neighbor algorithms in classification

In this paper, classification efficiency of the conventional K-nearest neighbor algorithm is enhanced by exploiting fuzzy-rough uncertainty. The simplicity and nonparametric characteristics of the conventional K-nearest neighbor algorithm remain intact in the proposed algorithm. Unlike the conventional one, the proposed algorithm does not need to know the optimal value of K. Moreover, the generated class confidence values, which are interpreted in terms of fuzzy-rough ownership values, do not necessarily sum up to one. Consequently, the proposed algorithm can distinguish between equal evidence and ignorance, and thus the semantics of the class confidence values becomes richer. It is shown that the proposed classifier generalizes the conventional and fuzzy KNN algorithms. The efficacy of the proposed approach is discussed on real data sets.