摘要
针对K-均值算法对初始值敏感和易陷入局部最优的缺点,提出了一种基于改进差分进化的K-均值聚类算法。该算法通过引入基于Laplace分布的变异算子和Logistic变尺度混沌搜索来增强全局寻优能力。实验结果表明,该算法能够较好地克服传统K-均值算法的缺点,具有较好的搜索能力,且算法的收敛速度较快,鲁棒性较强。
The conventional k-means algorithms are sensitive to the initial cluster centers, and tend to be trapped by local opti- ma. To resolve these problems, a novel k-means clustering algorithm using enhanced differential evolution technique is proposed in this paper. This algorithm improves the global search ability by applying Laplace mutation operator and variable-scale Logistic chaotic searching. Numerical experiments show that this algorithm overcomes the disadvantages of the conventional k-means al- gorithms, and improves search ability with higher accuracy, faster convergence speed and better robustness.
作者
高平
毛力
宋益春
GAO Ping, MAO Li2, SONG Yi-chun2 （1 Xinje Electronic Co., Ltd. , Wuxi 214000, China; 2 .Key Laboratory of Advanced Process Control for Light Industry （Minis- try of Education）, School of Internet of Things, Jiangnan University, Wuxi 214122, China）
出处
《电脑知识与技术》
2013年第8期5064-5067,共4页
Computer Knowledge and Technology
基金
轻工过程先进控制教育部重点实验室开放课题资助（江南大学）项目（APCLI1004）
