基于仿生优化算法的聚类改进算法  

Improved clustering algorithm based on bionic optimization algorithm

作　　者：覃承友 谢晓兰[1,2] 王悦悦 郭杨 QIN Cheng-you;XIE Xiao-lan;WANG Yue-yue;GUO Yang(Guangxi Key Laboratory of Embedded Technology and Intelligent System,Guilin University of Technology,Guilin 541006,China;College of Information Science and Engi-neering,Guilin University of Technology,Guilin 541006,China)

机构地区：[1]桂林理工大学广西嵌入式技术与智能系统重点实验室,广西桂林541006 [2]桂林理工大学信息科学与工程学院,广西桂林541006

出　　处：《桂林理工大学学报》2022年第1期223-228,共6页Journal of Guilin University of Technology

基　　金：国家自然科学基金项目(61762031);广西科技重大专项(桂科AA19046004);广西重点研发计划项目(桂科AB18126006)。

摘　　要：为提升聚类算法的聚类效果,采用仿生优化算法与k均值聚类算法相结合(BFOA-K)实现数据聚类。在聚类过程中,为解决k均值对于初始质心敏感以及容易陷入局部最优的问题,使用果蝇优化算法确定k均值聚类算法的质心,再使用k均值聚类算法进行数据聚类。针对果蝇优化算法对于飞行步长的影响,采用F分布动态改变步长,提升算法全局搜索能力。同时采用精英保留策略,提升果蝇种群的多样性,扩大了搜索范围和提升了搜索效率。利用4个UCI标准数据集对算法进行仿真实验,结果表明,本文提出的BFOA-K算法在各项聚类评估指标中都优于其余对比算法,提升了算法的收敛性,证明了算法的有效性和可行性。To improve the clustering of the clustering algorithm,the bionic optimization algorithm and the k-means clustering algorithm are combined to realize data clustering(BFOA-K).In the clustering process,the fruit fly optimization algorithm is used to determine the centroid of the k-means clustering algorithm,and the k-means clustering algorithm is used for data clustering.To solve the problem that k-means is sensitive to the initial centroid and easy to fall into the local optimum,and the influence of the fruit fly optimization algorithm on the flight step length,the F distribution is used to dynamically change the step length which improves the global search ability of the algorithm.The elite retention strategy is adopted to increase the diversity of the fruit fly population,expand the scope of the search and improve the search efficiency.Four UCI standard datasets are used to test the algorithm.The experiment shows that the algorithm is better than other comparison algorithms in all clustering evaluation indicators.The BFOA-K improves the convergence of the algorithm with effectiveness and feasibility.

关 键 词：仿生优化算法 K均值聚类 果蝇优化算法 F分布 精英保留策略 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]