[1]燕 存,吉根林.Item-Based并行协同过滤推荐算法的设计与实现[J].南京师大学报(自然科学版),2014,37(01):71.
　Yan Cun,Ji Genlin.Design and Implementation of Item-Based Parallel Collaborative Filtering Algorithm[J].Journal of Nanjing Normal University(Natural Science Edition),2014,37(01):71.
点击复制

Item-Based并行协同过滤推荐算法的设计与实现()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
第37卷
期数:
2014年01期
页码:
71
栏目:
计算机科学
出版日期:
2014-03-30

文章信息/Info

Title:
Design and Implementation of Item-Based Parallel Collaborative Filtering Algorithm
作者:
燕 存; 吉根林
南京师范大学计算机科学与技术学院,江苏 南京 210023
Author(s):
Yan Cun; Ji Genlin
School of Computer Science and Technology,Nanjing Normal University,Nanjing 210023,China
关键词:
推荐系统; 协同过滤; Hadoop; MapReduce
Keywords:
recommendaton system; collaborative filterinig; Hadoop; MapReduce
分类号:
TP311
文献标志码:
A
摘要:
基于协同过滤的推荐已成为推荐系统中广泛采用的推荐技术.由于应用中用户数目和商品条目的日益增长,在计算相似度和计算预测时,单机集中式计算已不能满足推荐系统实时性和可扩展性的要求.针对这一问题,设计并实现了Item-Based并行协同过滤推荐算法.该算法采用Hadoop的MapReduce与HDFS架构,可分为Map与Reduce两个过程.通过在Map和Reduce节点上的并行处理可提高算法的执行效率.实验结果表明,该算法可明显减少推荐时间,提高推荐实时性,获得良好的可扩展性.
Abstract:
Collaboration filtering has been widely used in recommender system.However,with the increase of user numbers and item numbers,the computing ability of single computer can not satisfy the requirement of real-time performance and the requirement of scalability when computing the similarity and prediction.To address the issue,this paper presents an Item-Based parallel collaborative filtering recommendation algorithm.Adopting the framework of MapReduce and HDFS in Hadoop,the parallel algorithm is divided into two procedures,which are Map and Reduce.Through the computation in parallel in respective Map and Reduce nodes,the algorithm performance is improved.The experimental results show that the proposed algorithm can decrease the recommend time and has a good real-time performance and scalability.

参考文献/References:

[1] Schafer J B,Konstan J A,Riedl J.E-commerce recommendation applications[J].Data Mining and Knowledge Discovery,2001,5(1/2):115-153.
[2]Linden G,Smith B,York J.Amazeon.com recommendations:item-to-item collaborative filtering[J].IEEE Internet Computing,2003,7(1):76-80.
[3]O’Conner M,Herlocker J.Clustering items for collaborative filtering[C]//Proceeding of the ACM SIGIR Workshop on Recommender System.California:UC Berkeley,1999:121-128.
[4]李忠俊,周启海,帅青红.一种基于内容和协同过滤同构化整合的推荐系统模型[J].计算机科学,2009,36(12):142-145.
[5]Lee J S,Jun C H,Kim S H.Mining changes in customer buying behavior for collaborative recommendations[J].Expert System with Applications,2005,29(3):700-704.
[6]Schafer J B,Frankowski D,Herlocker J,et al.Collaborative Filtering Recommender Systems[M].Berlin Heidelberg:Springer,2007:291-324.
[7]Sarwer B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms[C]//Proceeding of the 10th International Conference on World Wide Web.Hong Kong:ACM Press,2001:285-295.
[8]Zhao Z D,Shang M S.User-based collaborative-filtering recommendation algorithms on hadoop[C]//Third International Conference on Knowledge Discovery and Data Mining.Thailand:IEEE,2010:478-481.
[9]Borthakur D,Gray J,Sarma J S,et al.Apache hadoop goes realtime at Facebook[C]//Proceedings of the 2011 ACM SIGMOD International Conference on Management of Data.Athens:ACM,2011:1 071-1 080.
[10]Jiang J,Lu J,Zhang G,et al.Scaling-up item-based collaborative filtering recommendation algorithm based on hadoop[C]//2011 IEEE World Congress on Services(SERVICES).Washington:IEEE,2011:490-497.
[11]Dean J,Ghemawat S.MapReduce:simplified data processing on large clusters[J].Communications of the ACM,2008,51(1):107-113.
[12]Shvachko K,Kuang H,Radia S,et al.The hadoop distributed file system[C]//2010 IEEE 26th Symposium on Mass Storage Systems and Technologies.Nevada:IEEE,2010:1-10.
[13]Ahn H J.A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem[J].Information Sciences,2008,178(1):37-51.

相似文献/References:

[1]高 娜,杨 明.一种改进的结合标签和评分的协同过滤推荐算法[J].南京师大学报(自然科学版),2015,38(01):98.
　Gao Na,Yang Ming.An Improved Unifying Tags and Ratings Collaborative Filteringfor Recommendation System[J].Journal of Nanjing Normal University(Natural Science Edition),2015,38(01):98.

备注/Memo

备注/Memo:
收稿日期:2013-03-31.
基金项目:江苏省自然科学基金重点项目(BK2011005).
通讯联系人:吉根林,博士,教授,博士生导师,研究方向:数据挖掘技术及应用.E-mail:glji@njnu.edu.cn
更新日期/Last Update: 2014-03-30