摘要
针对传统数据挖掘算法只适用于小规模数据挖掘处理,由于数据规模不断增大,其存在计算效率低、内存不足等问题,文中将MapReduce用于数据挖掘领域,对大数据挖掘中的MapReduce进行了并行化改进,并设计相应的并行化实现模型,以期满足大数据分析需求,完成低成本、高性能的数据并行挖掘与处理。
The traditional data mining algorithm is only suitable for small-scale data mining and processing,and its disadvantages of low computational efficiency and insufficient memory are exposed gradually with the increase of data scale.MapReduce is used in the field of data mining to analyze the MapReduce parallelization improvement of the traditional data mining algorithms;and the corresponding parallelization implementation model is designed to meet the demand of big data analysis,and successfully complete the low-cost and high-performance data parallel mining and processing.
作者
吕国
肖瑞雪
白振荣
孟凡兴
LU Guo;XIAO Ruixue;BAI Zhenrong;MENG Fanxing(Modern Education Technology Center,Hebei University of Architecture,Zhangjiakou 075000,China)
出处
《现代电子技术》
北大核心
2019年第11期161-164,共4页
Modern Electronics Technique
基金
2018年河北省科学技术厅创新能力提升计划项目(184576131D)
2017年河北省高等学校科学技术研究项目(QN2017322)
2018年张家口市科学技术和地震局市级科技计划自筹经费项目(1821016B)~~
