计算机与数字工程2018，Vol.46Issue(3)：419-423,428,6.DOI:10.3969/j.issn.1672-9722.2018.03.001
VDOD:一种基于KD树的分布式离群点检测算法
VDOD:Distributed Outlier Detection Algorithm Based on KD-tree
摘要
Abstract
A new distributed outlier detection algorithm,VDOD is proposed for large data of the large amount of data and high dimensional characteristics.In the data preprocessing stage,a data partitioning method based on variance is proposed.KD tree is es?tablished in the process of partitioning,and the data are evenly distributed to each computing node through KD tree.In the outlier detection stage,batch filtering is performed by R tree.Finally,the validity of the VDOD algorithm is verified based on the real data set and the artificial data set.The experimental results show that compared with the existing algorithms,the algorithm can signifi?cantly improve the computational efficiency and significantly reduce the network overhead.关键词
分布式/离群点检测/大数据/KD树Key words
distributed/outlier detection/large data/KD tree分类
信息技术与安全科学引用本文复制引用
李子茂,骆庆,刘晶..VDOD:一种基于KD树的分布式离群点检测算法[J].计算机与数字工程,2018,46(3):419-423,428,6.基金项目
国家自然科学基金项目"多视图信息融合的乳腺肿块计算机辅助诊断关键技术研究"(编号:61302192) （编号:61302192）
国家科技支撑计划项目子课题"民族特色农产品多语言网络交易展示平台关键技术集成与应用示范"(编号:2015BAD29B01)资助. （编号:2015BAD29B01）
