摘要
云计算数据库用于存储海量数据,对其进行挖掘可以提高云计算数据库的并行调度能力。传统方法采用云计算数据库访问信道属性权重分配方法进行数据挖掘,数据挖掘性能较差。提出一种基于属性相关度估计的云计算数据库海量数据挖掘算法。建立云计算数据库的多层自回归矢量空间模型,在多层自回归矢量空间进行海量数据的特征分析和数据聚类,采用属性相关度估计方法提取海量数据的优化特征,实现数据挖掘算法改进。仿真结果表明,采用该算法进行云计算数据库数据挖掘,提高了数据挖掘的峰值聚焦性能和查准率,改善了数据库的优化访问和数据调度能力。
Cloud computing database is used to store huge amounts of data, and it can improve the parallel scheduling capability of cloud computing database. Traditional methods uses cloud computing database to access channel attribute weights allocation method for data mining, data mining performance is poor. A kind of massive data mining algorithm for cloud computing database based on attribute correlation estimation is put forward. The cloud computing database multiple regression vector space model is built and characteristic analysis and data clustering for huge amounts of data is carried out in that space. The attribute correlation estimation method is used to extract optimal characteristics of huge amounts of data, and the data mining algorithm is improved. The simulation results show that using this algorithm of cloud computing database data mining, the peak focusing performance and precision, and the database access optimization and data scheduling ability is improved.
出处
《控制工程》
CSCD
北大核心
2016年第6期956-960,共5页
Control Engineering of China
基金
浙江省教育厅课题(Y201432552)
关键词
云计算数据库
数据挖掘
属性相关度估计
Cloud computing database
data mining
attribute correlation estimation