摘要
K均值算法是以随机的方式选择初始聚类中心,这使得K均值算法容易陷入局部最优,收敛性能不稳定。针对这一缺陷,该文对K均值算法进行改进,提出一种逐步选择距离差异极大的个体作为初始聚类中心的算法。实验结果表明,改进后的算法收敛性能确实比K均值算法优越。
K-means algorithm chooses initial clustering centers randomly, which makes K-means algorithm easy to fall into local optimum and its convergence performance unstable. In order to overcome this shortcoming, an improved K-means algorithm has been proposed in this paper, which chooses gradually individuals with great distance difference as the initial clustering center. The experimental results show that the improved algorithm has better convergence performance than K-means algorithm.
出处
《科技资讯》
2019年第15期185-187,共3页
Science & Technology Information
关键词
K均值算法
聚类中心
局部最优
收敛性
K-means algorithm
Clustering center
Local optimum
Convergence
