A Weighted K-means Algorithm Based on Differential Evolution

摘要

The k-means algorithm is constrained by the initial clustering centers and abnormal data, and unstable clustering results are easy to occur. In order to solve this problem, this dissertation firstly analyzes the research status of the differential evolution algorithm, introduces the basic idea and advantages of the differential evolution algorithm, proposes a weighted evolution algorithm based on differential evolution, adopts the differential evolution algorithm with strong global search ability, Initial clustering centers. According to the different influence degree of samples on clustering analysis, weights are introduced to design a weighted Euclidean distance to reduce the adverse effects of uncertainties such as outliers and to obtain a stable clustering result The experimental results show that the initial clustering center selected by the algorithm is closer to the final clustering center, and the computational efficiency of the algorithm is improved while ensuring the clustering accuracy.