摘要
针对传统算法在挖掘负载平衡数据时,常常会出现挖掘效率低、误差高等问题,提出基于耦合度量的负载平衡大数据挖掘方法.在对耦合度量算法分析后,利用K-tras分割聚类算法不断更新聚类中心,完成负载平衡大数据的聚类;计算负载平衡数据的最优分类面和量化后的矢量轨迹,完成数据挖掘.实验结果表明,所提方法取得了理想的挖掘效率、查全率及较低的挖掘误差.
Because the traditional algorithms often have the problems of low mining efficiency and high error when mining load balancing data,a load balancing big data mining method based on coupling measurement is proposed.After analyzing the coupling measurement algorithm,K-tras segmentation clustering algorithm is used to update the clustering center constantly,and the clustering of load balancing big data is completed.The optimal classification surface and quantized vector trajectory of load balancing data are calculated to complete data mining.By setting up an experimental platform and carrying out comparative simulation experiments,the results show that the proposed method achieves the best mining efficiency and recall ratio,and the lowest mining error.
作者
蔡燕萍
CAI Yan-ping(Department of Information and Technology,Xiamen Xingcai Vocational and Technical College,Xiamen 361024,Fujian,China)
出处
《兰州文理学院学报（自然科学版）》
2023年第2期40-44,共5页
Journal of Lanzhou University of Arts and Science(Natural Sciences)
基金
福建省教育厅中青年教师教育科研项目(JZ180724)。
关键词
挖掘效率
矢量轨迹
数据挖掘
聚类中心
耦合度量
mining efficiency
vector trajectory
data mining
cluster center
coupling measurement
