K-NN algorithm based on neural similarity

The aim of this paper is to present a k-nearest neighbour (k-NN) classifier based on a neural model of the similarity measure between data. After a preliminary phase of supervised learning for similarity determination, the neural similarity measure is used to guide the k-NN rule. Experiments on both synthetic and real-world data show that the similarity-based k-NN rule outperforms the Euclidean distance-based k-NN rule.