关键词: 聚类, K均值算法, 初始聚类中心, 高密度区域

Abstract: The initial clustering center of the traditional K-means algorithm was generated randomly from the data set, and the clustering result was unstable. An improved K-means algorithm based on density algorithm optimizing initial clustering center was proposed, which selected the furthest mutual distance k points in highdensity region as the initial centers. The experimental results demonstrate that the improved K-means algorithm can eliminate the dependence on the initial cluster center, and the clustering result has been greatly improved.

Key words: clustering, K-means algorithm, initial clustering center, high-density area