首页 | 本学科首页   官方微博 | 高级检索  
     

网络电视推荐系统框架及协同过滤算法的研究
引用本文:黄乔,尹四清.网络电视推荐系统框架及协同过滤算法的研究[J].电视技术,2013,37(9).
作者姓名:黄乔  尹四清
作者单位:中北大学电子与计算机科学技术学院,山西太原,030051
摘    要:针对网络电视推荐系统中通常采用的协同过滤推荐算法的不足,提出了一种将聚类、用户相似—信任关系和项目属性关系相组合的协同过滤推荐技术.该组合推荐技术首先通过聚类分析缩小用户的有效搜索范围,其次通过引入信任关系来提高推荐的准确性,从而为目标用户提供更好的推荐结果.经过实验表明,该算法提高了推荐质量.

关 键 词:网络电视  推荐系统  协同过滤  聚类  信任
收稿时间:2012/12/19 0:00:00
修稿时间:1/5/2013 12:00:00 AM

Network TV Recommended System Framework and Research of Collaborative Filtering Algorithm
huangqiao and yinsiqing.Network TV Recommended System Framework and Research of Collaborative Filtering Algorithm[J].Tv Engineering,2013,37(9).
Authors:huangqiao and yinsiqing
Affiliation:North University of China,North University of China
Abstract:In view of the Insufficient of network TV recommender system which usually uses collaborative filtering recommendation algorithms, put forward a kind of combination of recommendation algorithm which is based on Clustering Similarty-Trust and Project properties relationship. Firstly,the combination of recommendation algorithm reduced user's effective range through cluster analysis,then improve the accuracy of recommendation through the trust mechanism, so as to provide better recommendation results for target users. The experimental results show that, the algorithm improves the quality of recommendation.
Keywords:Network TV  recommender systems  collaborative filtering  clustering  trust  
本文献已被 万方数据 等数据库收录!
点击此处可从《电视技术》浏览原始摘要信息
点击此处可从《电视技术》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号