摘要
现有大数据集中碎片数据实时标记方法存在标记实时性差、鲁棒性差的问题,为了解决上述问题,提出基于Java的大数据集中碎片数据实时标记方法。提取大数据中碎片数据,以碎片数据特征为基础创建最优数据集合树,完成碎片数据的集合,得到集合碎片数据,利用线性函数转换方法处理集中碎片数据。选取适当的核函数,确定标记因子,以确定的标记因子为依据,基于Java平台编写集中碎片数据实时标记程序,实现大数据集中碎片数据的实时标记。实验结果表明,提出的大数据集中碎片数据实时标记方法极大的提升了标记实时性与鲁棒性,充分说明提出的大数据集中碎片数据实时标记方法具备更好的性能。
The existing real time marking methods for fragmented data in large data sets have the problems of poor real time marking and poor robustness.To solve these problems,a real time marking method for fragmented data in large data sets based on Java is proposed.The fragmentation data is extracted from large data,the optimal data set tree is created based on the characteristics of fragmentation data,the fragmentation data set is completed,and the fragmentation data set is obtained.The fragmentation data is processed by linear function transformation method.Choose the appropriate kernel function,determine the marking factor,and write a real time marking program based on Java platform to realize the real time marking of fragmentation data in large data sets.The experimental results show that the proposed real time marking method greatly improves the real time performance and robustness of the marking,which fully demonstrates that the proposed real time marking method for large data sets has better performance.
作者
王岩
WANG Yan(Information Technology College,Shenyang Institute of Technolog,Fushun 113122,China)
出处
《电子设计工程》
2020年第9期46-49,53,共5页
Electronic Design Engineering
关键词
JAVA
大数据
集中碎片数据
实时标记
Java
big data
centralized fragmentation data
real time markup
