基于改进磷虾群算法的K-means算法  

K-means Algorithm Based on Improved Krill Herd Algorithm

作　　者：刘唐 周炜 李志鹏 权文 LIU Tang;ZHOU Wei;LI Zhipeng;QUAN Wen(Air Force Engineering University,Xi'an 710051,China;Xi'an Institute of Finance and Economics,Xingzhi School,Xi'an 710038,China;Unit 75837 of PLA,Guangzhou 510630,China)

机构地区：[1]空军工程大学 [2]西安财经学院行知学院 [3]中国人民解放军75837部队

出　　处：《探测与控制学报》2019年第1期76-81,共6页Journal of Detection & Control

基　　金：国家自然科学基金项目资助(61503407)

摘　　要：针对磷虾群算法易陷入局部最优、搜索能力弱及K-means算法易受初始聚类中心选择影响等问题,提出一种基于改进磷虾群算法的K-means算法。该算法通过混沌初始化、动态分群、精英引领和随机变异等策略改进磷虾群算法,并引入最佳聚类数自适应机制,提高了算法的综合寻优能力。实验通过6种基准函数检验了改进磷虾群算法的有效性,用UCI机器学习数据集及人造数据集测试验证了基于改进磷虾群算法的K-means算法的性能。验证结果表明,改进磷虾群算法在保证较快收敛速度的基础上提升了全局寻优能力,与其他算法相比,该算法各方面性能显著提升。In order to solve the problem of weak searching ability of standard krill herd algorithm ,which is vulnerable to the local optimum,and the K-means algorithm is susceptible to the selection of the initial clustering center,a K-means algorithm based on the improved krill herd algorithm was proposed.The algorithm improved the krill herd algorithm through chaos initialization,dynamic clustering,elite leading and random mutation strategy,and the adaptive mechanism of optimal cluster number was introduced,the comprehensive optimization ability of the algorithm was improved.The validity of the algorithm was verified by six benchmark functions.The performance of the algorithm was validated by UCI machine learning data set and artificial data set test.The verification results showed that the improved krill herd algorithm could improve the global optimization ability on the basis of ensuring the fast convergence speed,and the performance of this algorithm was significantly improved compared with other algorithms.

关 键 词：磷虾群算法 聚类算法 精英引领 最佳聚类数 动态分群 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]