作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2015, Vol. 41 ›› Issue (1): 245-250. doi: 10.3969/j.issn.1000-3428.2015.01.046

• 多媒体技术及应用 • 上一篇    下一篇

社交网络服务中的多维空间视频推荐算法

周娇a,霍欢a,b   

  1. 上海理工大学 a.光电信息与计算机工程学院;b.上海现代光学系统重点实验室,上海 200093
  • 收稿日期:2014-01-13 修回日期:2014-03-18 出版日期:2015-01-15 发布日期:2015-01-16
  • 作者简介:周 娇(1988-),女,硕士,主研方向:数据挖掘,自然语言处理;霍 欢,副教授、博士。
  • 基金资助:
    国家自然科学基金资助项目(61003031)

Multi-dimensional Space Video Recommendation Algorithm in Social Network Service

ZHOU Jiaoa,HUO Huana,b   

  1. a.School of Optical-electrical and Computer Engineering; b.Shanghai Key Laboratory of Modern Optical System,
    University of Shanghai for Science and Technology,Shanghai 200093,China
  • Received:2014-01-13 Revised:2014-03-18 Online:2015-01-15 Published:2015-01-16

摘要: 视频推荐作为一项帮助用户迅速找到其最感兴趣视频的关键技术,是社交网络服务中比较重要的研究内容之一。传统推荐算法未能充分利用视频社会化网站中的多维信息,会导致冷启动和数据稀疏的问题。为此,提出一种社交网络服务中的多维空间视频推荐算法。综合分析构成视频社会化网络的多维信息源要素,在此基础上,通过构建多维聚类空间,进而实现基于多维聚类空间的视频推荐算法,利用构成视频社会化网络的多维信息源要素,为视频的个性化推荐提供信息来源,以解决冷启动和数据稀疏问题。实验结果表明,该算法在视频推荐准确度方面相对于传统视频推荐算法有明显提高。

关键词: 视频推荐, 社交网络服务, 多维空间, 属性相似性, 内容相似性, 社交关联性

Abstract: The video recommendation,as a key enabling technology to provide users with the most interested and relevant videos,is one of the most important research topics in Social Network Service(SNS).This paper presents a multi-dimensional space based video recommendation algorithm in SNS,againsts the cold start and sparse caused by the traditional recommendation algorithms ignoring multidimensional information.By analyzing the multidimensional information sources in the video social networking sites,the paper imports various elements into the video recommendation,to construct multi-dimensional space by clustering and implement recommendation for user-generated videos based on multi-dimensional clustering space.The algorithm takes full use of the multidimensional information elements which constitute the video social networking providing a rich source of information for video personalized recommendation,and solves the problem of cold start and sparse data.Experimental results demonstrate the effectiveness of the multi-dimensional space based video recommendation algorithm,which achieves a significantly higher recommendation accuracy than the traditional video recommendation algorithms.

Key words: video recommendation, Social Network Service(SNS), multi-dimensional space, attribute similarity, content similarity, social correlation

中图分类号: