欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对模糊C-均值（FCM）算法必须预先给定聚类数c和容易陷入局部极小的缺点,提出了融合遗传算法和粒子群算法的GA-PSO-FCM算法。遗传算法（GA）嵌套在FCM算法的外层,用于自动寻找最优聚类数,并把有效性准则函数作为其适应度函数;粒子群（PSO）算法嵌套在FCM算法的内层,用于优化类中心向量,提高算法的全局搜索能力。最后,运用GA-PSO-FCM算法对Iris data、Wine data、Zoo data、WPBC data和WDBC data进行仿真实验,并与基于有效性准则函数改进的FCM算法、GA-FCM算法的仿真结果进行比较,表明GA-PSO-FCM算法能在预先未知聚类数的情况下,提高分类结果的精确性和稳定性。
英文摘要：
This paper proposes a GA-PSO-FCM algorithm that integrates GA with PSO to overcome the shortcomings that FCM algorithm needs a given cluster number c in advance and is easy to get into local minimum.GA is embedded as the outer layer of FCM algorithm to search for the optimal cluster number,and takes validity index as its fitness function;PSO is embedded as the inner layer of FCM algorithm to optimize the cluster center vector to improve the ability of global search.The GA-PSO-FCM algorithm was tested with Iris data,Wine data,Zoo data,WPBC data and WDBC data.The results were compared with simulation result of FCM and GA-FCM.It shows that GA-PSO-FCM algorithm may improve the accuracy and stability of clustering without knowing the cluster number in advance.
同期刊论文项目
同项目期刊论文
Simulation on time-of-use pricing policy in China urban natural gas pipeline network based on multi-
An optimal time-of-use pricing for urban gas: A study with a multi-agent evolutionary game-theoretic
A hybrid coding SA method for multi-item capacity-constrained production and delivery scheduling pro
Application of Hybrid GA-SA Heuristics for Single-job Production-delivery Scheduling Problem with In
Using Multi-Agent Method for Application of TOU Pricing Policy in China Urban Natural Gas Pipeline N
Comparative study on economic contribution rate of education of China and foreign countries based on
期刊信息
位置：