王硕,孙光明,邹静昭,李伟生.基于用户推荐影响度的并行协同过滤算法[J].计算机科学,2017,44(9):250-255, 271
基于用户推荐影响度的并行协同过滤算法
Parallel Collaborative Filtering Algorithm Based on User Recommended Influence
投稿时间:2016-08-13  修订日期:2016-11-09
DOI:10.11896/j.issn.1002-137X.2017.09.047
中文关键词:  推荐影响度,推荐新颖度,兴趣重合度,MapReduce并行化
英文关键词:Recommendation influence degree,Recommendation novelty degree,Interest coincidence degree,MapReduce paralleliation
基金项目:本文受河北省高等学校科学技术研究重点项目(ZD2014061),青年基金项目(QN2016108)资助
作者单位E-mail
王硕 河北科技大学信息科学与工程学院 石家庄050035  
孙光明 北京交通大学计算机与信息技术学院 北京100004 sungmwhu@whu.com 
邹静昭 河北中医学院公共课教学部 石家庄050200  
李伟生 北京交通大学计算机与信息技术学院 北京100004  
摘要点击次数: 90
全文下载次数: 48
中文摘要:
      基于共同评分与项目全集的相似度未甄别近邻的推荐影响力,导致推荐质量低,可扩展性差。为此,提出了一种基于推荐影响度的并行协同过滤算法。该算法通过非共同评分项目、共同评分项类以及用户访问次数来计算用户推荐新颖度与兴趣重合度以度量用户推荐能力,并融入相似性计算来抑制相似度高但推荐力不强的用户,避免在项目全集上计算相似度,从而提高推荐质量;通过MapReduce并行化,使其具备良好的实时性和可扩展性。实验结果表明,该算法在海量数据集上的推荐质量更高,可扩展性更强。
英文摘要:
      The similarity based on common scores and full item sets has failed to identify the nearest neighbor recommendation influence,which brings about lower recommend quality and poor scalability.Through non-common rating items,common score item categories and user visited times,this paper proposed a parallel collaborative filtering algorithm based on user recommendation influence.It computes the user recommended novelty degree and interest coincidence to measure user recommendation influence ability.By adding it to calculate similarity,the algorithm can effectively restrain the highly recommended users with high similarity,avoid similarity computation on full item sets and improve the quality of recommendation. Further more,by using MapReduce parallelization,this algorithm has good real-time performance and scalability.The experimental results show that the parallel algorithm is of higher recommendation quality and better scalability on big data.
查看全文  查看/发表评论  下载PDF阅读器