The optimized evidence k-Nearest Neighbor based on FOA under the hesitant fuzzy environment and its application in classification

摘要

The k-Nearest Neighbor (k-NN) is one of the simplest intelligent algorithms in the field of pattern recognition and classification. The increasing complexity of practical applications brings more uncertainty and fuzziness. In this paper, we take advantage of the Dempster-Shafer evidence theory (D-S evidence theory) and the hesitant fuzzy set (HFS) in depicting uncertain preference and information, and develop the evidence k-Nearest Neighbor (Ek-NN) under the hesitant fuzzy environment. The fruit fly optimization algorithm (FOA) is adopted to determine the most appropriate value of k in Ek-NN, and a specific implementation process of the optimized Ek-NN based on FOA is also provided. Moreover, two numerical examples about classification problems are presented to evaluate the performance of the proposed method. Comparative analysis and sensitivity analysis are further conducted to illustrate the advantages of the optimized Ek-NN based on FOA under the hesitant fuzzy environment.