一种数据挖掘算法在Hadoop集群上的研究与实现  

Research and Implementation of a Data Mining Algorithm on Hadoop Cluster

作　　者：孙承秀 乔永峰[1] 

机构地区：[1]郑州电力职业技术学院

出　　处：《工业控制计算机》2019年第10期89-90,93,共3页Industrial Control Computer

摘　　要：近年来,采集和处理大容量数据,从大量数据中提取出有价值数据的技术在不断提升。用户对庞大的数据迅速获取、发掘和分析处理,洞察有价值的数据、获得商业价值、帮助用户识别、实时决策、预测趋势并做出响应是数据挖掘的本质。根据用户的网络行为确定用户的品位与爱好,推荐给用户可能感兴趣的物品、新闻或好友,向企业提供实时状态更新的依据等,是数据挖掘的目标。协同过滤算法是数据挖掘的案例,也是最为热门的机器学习算法。阐述了基于物品的协同过滤算法在虚拟机Hadoop集群上的实现,探讨大数据挖掘技术的实现过程。The goal of data mining is what recommend the items,news or friends that may be of interest to the user by the user's hobbies and network behavior.Provide the basis for real-time status updates to the enterprise,etc.CollaborationFilter is a case of data mining and the most popular machine learning algorithm.The technology of collecting and processing large-capacity data and extracting valuable data from a large amount of data has been improved continuously in recent years.Rapidly acquiring,discovering and analyzing big data,gaining business value,helping users identify,making real-time decisions,predicting trends responding by valuable data,is the essence of data mining.This paper describes the implementation of the Item CF on the virtual machine Hadoop cluster,and discusses the implementation process of big data mining technology.

关 键 词：虚拟机Hadoop集群 基于物品的协同过滤算法 数据挖掘 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]