摘要
传统的模糊C-均值聚类算法存在对初始聚类中心选择与噪声数据敏感,容易使目标函数陷入局部最优的问题,以及标准人工蜂群算法局部搜索能力及开发能力不强的缺点。针对这些问题,引进差分进化的思想改进人工蜂群算法并对跟随蜂的搜索行为进行更准确的描述,结合模糊C-均值聚类算法具有收敛速度快、易于实现且局部搜索能力较强的优点,提出一种基于模糊C-均值的改进人工蜂群聚类算法以提高聚类的性能。实验结果表明,该算法相对于传统FCM聚类算法,其准确率和抗噪性有所提高,聚类效果更好。
Due to the issues of traditional fuzzy C-means（ FCM） clustering algorithm,which is sensitive to the initial selection of the center and noise data,and easy to make an objective function into local optimum,and the disadvantages of weak local search ability and development capability in standard artificial bee colony algorithm,this paper modified artificial bee colony algorithm inspired by the thought of differential evolution and made a more accurate description of searching behavior of onlooker bees. Fuzzy C-means clustering algorithm has advantages of fast converges,the ability of local search and is easy to implement. Combination of them can improve the performance of clustering. The experiment shows that compared with traditional FCM algorithm,the algorithm further improves the accuracy and noise immunity and has better clustering results.
出处
《计算机应用研究》
CSCD
北大核心
2016年第5期1342-1345,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(61303074
61309013)
河南省科技攻关计划资助项目(12210231003
13210231002)
关键词
人工蜂群算法
模糊C-均值
聚类分析
差分进化
搜索方程
artificial bee colony（ABC） algorithm
fuzzy C-means
clustering analysis
differential evolution
search equation
