A generalized fuzzy k-nearest neighbor regression model based on Minkowski distance

Abstract The fuzzy k-nearest neighbor (FKNN) algorithm, one of the most well-known and effective supervised learning techniques, has often been used in data classification problems but rarely regression settings. This paper introduces a new, more general model. Generalization is based on usage Minkowski distance instead usual Euclidean distance. not optimal choice for practical problems, better results can be obtained by generalizing this. Using allows proposed method to obtain reasonable nearest neighbors target sample. Another key advantage this that are weighted weights their similarity sample, leading accurate prediction through average. performance tested with eight real-world datasets from different fields benchmarked k -nearest three other state-of-the-art methods. Manhattan distance- distance-based FKNNreg methods also implemented, compared. empirical show (Md-FKNNreg) outperforms benchmarks good algorithm problems. In particular, Md-FKNNreg model gave significantly lowest overall average root mean square error (0.0769) all used. As special case distance, yielded conditions achieved best datasets.