摘要
针对K-means聚类算法易陷入局部极小的问题,利用动态隧道算法在解决全局最优化问题中的有效性,将算法中的动态隧道过程引入到K-means聚类算法中,提出了一种基于动态隧道算法的K-means聚类算法。该算法在K-means聚类算法寻优得到的局部极小值基础上,利用动态隧道过程寻找更小的能量盆地,再将其值提交给K-means聚类算法进行迭代寻优,重复该过程,直到找到全局最小值。理论分析和仿真实验证明,该算法的聚类效果要优于K-means聚类算法。
K-means clustering algorithm itself is a global optimization problem, whose objective function has multiple local minima and only one global minimum. The algorithm is apt to fall into local minimum points through its iteration process. Dynamic tunneling algorithm is especially developed for the global optimization problem. On the basis of the local minimum got by dynamic optimization process, dynamic tunneling process makes use of tunneling to skip the local minimum point and search a lower energy valley that in essence means a lower value point than the local minimum point, thus a new initial point obtained can further be transferred to dynamic optimization process to be optimized. The use of the validity of dynamic tunneling algorithm, dynamic tunneling process of the algorithm is introduced into K-means clustering algorithm, and a novel K-means clustering algorithm based on dynamic tunneling system is presented in this paper. On the basis of local minimum got by K-means clustering algorithm, dynamic tunneling process is used to search a lower energy valley, then the value is submitted to K-means clustering algorithm for iterative optimization. The process is unceasingly repeated until the global minimum point is found. Both the theory analysis and simulation experiment results show that the presented algorithm in this paper is superior to K-means clustering algorithm.
出处
《重庆师范大学学报（自然科学版）》
CAS
2009年第1期73-77,共5页
Journal of Chongqing Normal University:Natural Science
基金
国家自然科学基金(No.10171118)
重庆市教委科学技术研究项目(No.KJ060818)
运筹学与系统工程重庆市市级重点实验室开放课题(No.YC200804)
