摘要
海量数据挖掘算法研究是当前数据挖掘研究领域的热点问题。以网络终端海量数据分布处理及软硬件资源共享为基础,以两变量线性关联效应测度学习算法设计为例,设计出基于超海量数据各终端数据分布处理的学习算法,并运用实验数据验证了该学习算法的有效性。该学习算法设计为海量数据云计算提供了应用思路。
The study of massive data mining algorithm is the research hotspot. On the basis of network terminal hardware and software information distributed processing as well as hardware and software resource sharing, taking two variables and linear correlation effect measure learning algorithm design as an example, the authors put forward learning algorithm based on massive data and the terminal data distribution processing. Experiments show that the learning algorithm is effective, which provides thought for cloud comouting of massive data.
出处
《统计与信息论坛》
CSSCI
2012年第7期19-22,共4页
Journal of Statistics and Information
基金
全国统计科学重点研究项目<统计数据质量多维系数诊断方法理论与应用研究>(2010LB35)
陕西省社会科学基金<金融危机对在陕外商投资的影响及应对措施研究>(09E20)
关键词
海量数据
资源共享
效应测量
关联算法
massive data
resource sharing
effect measure
correlation algorithm