| 一种基于MapReduce的关联规则挖掘算法 |
| |
| 引用本文： | 周国军.一种基于MapReduce的关联规则挖掘算法[J].玉林师范学院学报,2014(5):128-134. |
| |
| 作者姓名： | 周国军 |
| |
| 作者单位： | 玉林师范学院数学与信息科学学院,广西玉林537000 |
| |
| 基金项目： | 广西教育厅2014年度广西高校科学技术研究项目（项目编号：LX2014300） |
| |
| 摘 要： | 本文从减少I/O时间的角度出发,结合云计算Hadoop平台的Map Reduce模型,提出了一种基于Map Reduce的关联规则挖掘算法.算法采用幂集计算候选项集,采用Map Reduce模型在多个节点上并行找出所有频繁项集,只需要扫描事务数据库1次.实验结果表明：在事务的平均项长较小的情况下,算法具有很好的加速比和数据规模增长性.
|
| 关 键 词： | 关联规则 MapReduce 云计算 Hadoop |
An Algorithm of Mining Association Rules Based on MapReduce |
| |
| ZHOU Guo-jun.An Algorithm of Mining Association Rules Based on MapReduce[J].Journal of Yulin Teachers College,2014(5):128-134. |
| |
| Authors: | ZHOU Guo-jun |
| |
| Institution: | ZHOU Guo-jun (College of Maths ＆ Information Science, Yulin Normal University, Yulin, Guangxi 537000) |
| |
| Abstract: | From the viewpoint of reducing I/O time, according to Map Reduce of Hadoop platform of cloud computing, this paper presents an algorithm for mining association rules based on Map Reduce. The algorithm uses power set to compute candidate itemsets, and finds all frequent itemsets in parallel with Map Reduce model, which scan the transaction database only once. Experimental result shows that the algorithm can achieve a good speedup and data sizeup under the condition that the length of itemset in transaction is not very longer. |
| |
| Keywords: | association rules Map Reduce cloud computing Hadoop |
| 本文献已被 维普 等数据库收录！ |
