摘要
通过对当前有代表性的离群数据检测方法的分析和比较,总结了各方法的特性及优缺点.针对大数据的数据量大、维数高的特性,分析了离群点检测方法的改进策略,并以T-ODCD算法和AROD算法为例,进一步说明离群点检测改进策略.
The paper compared and analyzed major outlier detection method and their features and merit and demerit were summarized. In addition,in view of the large amount of data and high dimension of the big data,improvement strategies of outlier detection method were analyzed. Improvement strategies of outlier detection were further illustrated by T-ODCD and AROD algorithms.
出处
《江西师范大学学报（自然科学版）》
CAS
北大核心
2014年第5期454-458,495,共6页
Journal of Jiangxi Normal University(Natural Science Edition)
基金
国家社科基金教育学青年课题"教育虚拟社区的群集智能化构建方法研究"(CCA110109)资助项目
关键词
大数据
离群点检测方法
改进策略
big data
outlier detection method
improvement strategies
