基于Hadoop的推荐算法实现  

Implementation of Recommendation Algorithm Based on Hadoop

作　　者：黄鑫 景亮[1] HUANG Xin JING Liang(School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013 China)

机构地区：[1]江苏大学电气信息工程学院,江苏镇江212013

出　　处：《自动化技术与应用》2017年第10期65-67,99,共4页Techniques of Automation and Applications

摘　　要：为了解决在现今的大数据环境下从海量数据中进行快速推荐的问题,本文通过搭建Hadoop平台,利用MapReduce模型实现了基于内容推荐算法的并行化,能够有效的从大量数据中为用户做出推荐。实验证明,同传统串行计算相比,基于MapReduce模型的算法可以有效降低时间复杂度,并且表现出很好的性能。In order to solve how to make quick recommendations in today＇ s big data environment, this text establishes the Hadoop platform, using the MapReduce model to realize the parallelization of recommendation algorithm based on content. It can effectively make recommendation for customers from massive data. The experimental results show that MapReduce algorithm can effectively reduce the time complexity compared with the traditional serial computation, and it shows good performance.

关 键 词：HADOOP MAPREDUCE 推荐算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]