Evolutionary fuzzy k-nearest neighbors algorithm using interval-valued fuzzy sets.

•EF-kNN-IVFS, a new fuzzy nearest neighbor classification algorithm based on interval-valued fuzzy sets and evolutionary algorithms is presented.•Interval-valued fuzzy sets provide a way of representing several configurations for the parameters of fuzzyKNN.•Those configurations are set up in an adaptive way: an evolutionary method (CHC) searches for the best possible configuration according to the training data available.•An extensive experimental study demonstrates the good behavior of EF-kNN-IVFS, when compared with other algorithms of the state of the art.