摘要
大数据中的超大数据量级和数据集的复杂化需要专业的分析工具进行分析。本文提出了一种相关分析技术,用于大数据内容分析。本文所提相关分析技术利用了超多自变量的最佳线性拟合回退。为验证所提算法性能,本文采用了离散点绘图和最优线性回退方式进行验证。验证结果证明所提算法能够清晰描绘出大数据所含信息特征轮廓,并发现众多变量与内容之间的关联性。
Specific tools are requested for analyzing the huge amount of big data as well as the alter complexity of big data. A correlation analysis method of big data for Machine Learning (CAM) is proposed in this paper. CAM is charted for big data analysis. CAM utilizes ultra many variables to obtain the best linear regression. To test CAM, the distributed plotting and optimal linear regression is used. Test results show that CAM can clearly describe the feature of big data and find out the relation of variables and contents.
出处
《激光杂志》
北大核心
2016年第8期145-148,共4页
Laser Journal
基金
北京科技计划项目(14ZCDGSF0004)
关键词
定量分析
相关分析
大数据
quantities analysis
correlation analysis
big data