最小化误差平方和k-means初始聚类中心优化方法-计算机工程与应用.PDF

48 2018 ，54（15） Computer Engineering and Applications 计算机工程与应用 最小化误差平方和k-means 初始聚类中心优化方法 用 周本金，陶以政，纪 斌，谢永辉 应 ZHOU Benjin, TAO Yizheng, JI Bin, XIE Yonghui 与 中国工程物理研究院 计算机应用研究所，四川 绵阳 621900 程 Institute of Computer Application, China Academy of Engineering Physics, Mianyang, Sichuan 621900, China g 工 r o ZHOU Benjin, TAO Yizheng, JI Bin, et al. Optimizing k-means initial clustering centers by minimizing sum of . j squared error. Computer Engineering and Applications, 2018, 54 （15）：48-52. 机 a 算 e Abstract ：Traditional k-means algorithm is sensitive to initial clustering centers and isolated points, based on the principal c . of minimizing the sum of squared error to the most extent, an optimized k-means method is presented on selecting initial 计 w clustering centers. At the phase of initial selecting clustering centers, when adding a clustering point each time, compute reduced sum of squared error of each point and select the point that can maximize the square of the reduced error. Using w real datasets and compared with the results of other algorithms, the experimental results show the number of iteration is w reduced on selecting initial clustering centers and the quality of clustering is improved. Besides, artificial dataset demon- strates the method is much less sensitive to isolated points. Key words ：clustering; k-means algorithm; sum of squared error; isolated points 摘 要：传统的k-均值算法对初始聚类中心和孤立点敏感，文中以最大程度地减少误差平方和为基本思想，提出一