期刊文献+

基于活跃度指数的标签相关性判断研究 被引量:11

Study on Social Tags Relevance Judgment Based on Activity Index
原文传递
导出
摘要 [目的/意义]借鉴活跃度指数的设计思想,提出一种新的标签相关性判断策略,以改善标签相关性判断的效果和策略的通用性。[方法/过程]结合标签相关性判断的特点对活跃度指数的计算方法进行改造,进而提出一种基于多次活跃度指数迭代的标签相关性判断策略,并以社会化标注社区"豆瓣电影"的67 5351位用户的标签数据为例进行实验,以验证策略的效果。[结果/结论]实验结果显示,该策略的召回率为79.6%,准确率为93.3%,均较为理想,明显优于常用的Top-N策略。同时,该策略的通用性较好,适用于视频、音频、文本等各类型媒体。因此,该策略能够较好地解决标签的相关性判断问题。 [ Purpose/significance]The purpose of this paper is to improve both the effect and generality of social tags relevance judgment strategies. [ Method/process] On the basis of adjusting the calculation method of activity index by considering the characters of social tag relevance judgment, this paper proposes a tag relevance judgment algorithm based on multiple iteration of activity index, and verifies it by experiment based on 675351 Douban Movie users' social tagging data. [ Result/conclusion] The results show that this tactic performs well whose recall is 79.6% and precision is 93.3 % , and it is obviously better than Top-N social tags relevance judgment strategy. Moreover, it has good universality, and is suitable for various media such as video, audio, txt etc. Thus, it can be concluded that this tactic can solve the problem of tag relevance judgment well.
作者 林鑫 周知
出处 《图书情报工作》 CSSCI 北大核心 2015年第9期97-103,共7页 Library and Information Service
基金 教育部人文社会科学青年基金项目"社会网络环境下信息内容主题挖掘与语义分类研究(项目编号:13YJC870008)" 国家自然科学青年基金项目"社会网络环境下基于用户-资源关联的信息推荐研究(项目编号:71303178)"研究成果之一
关键词 活跃度指数 社会化标签 标签相关性 activity index social tag taga relevance
  • 相关文献

参考文献26

  • 1Xu Hao,Wang Jingdong,Hua Xian-Sheng, et al. Tag refinement by regularized LDA[ C ]//Proceedings of the 17th ACM International Conference on Multimedia. New York : ACM, 2009 : 573 - 576.
  • 2Li Xirong, Snoek C G M, Worring M. Learning social tag rele- vance by neighbor voting[ J ]. IEEE Transactions on Multimedia, 2009, 11(7): 1310-1322.
  • 3Lee S, De Neve W, Ro Y M. Visually weighted neighbor voting for image tag relevance learning [ J ]. Multimedia Tools and Applica- tions, 2013.72(2) : 1363 -1386.
  • 4李锡荣,许洁萍,薛盛博,杨刚.基于软近邻投票的图像标签相关性计算[J].计算机学报,2014,37(6):1365-1371. 被引量:4
  • 5Marinho L B, Schmidt -Thieme L. CollaboratNe tag recommenda- tions[ C]//Data Analysis, Machine Learning and Applications. Berlin : Springer Berlin Heidelberg, 2008 : 533 - 540.
  • 6吴思竹.社会标注系统中标签推荐方法研究进展[J].图书馆杂志,2010,29(3):48-52. 被引量:14
  • 7Oliveira B, Calado P, Pinto H S. Automatic tag suggestion based on resource contents [ C ]//Knowledge Engineering : Practice and Patterns. Berlin : Spriuger Berlin Heidelberg, 2008 : 255 - 264.
  • 8Li Xirong, Snoek C G M, Worring M. Unsupervised multi-feature tag relevance learning for social image retrieval [ C ]//Proceedings of the ACII International Conference on image and ideo Retrieval. New York: ACM, 2010:10-17.
  • 9Gao Yue, Wang Meng, Luan Huan-Bo, et al. Tag-based social im- age search with visual-text joint hypergraph earning[ C ]//Proceed- ings of the 19th ACM international conference on Multimedia. New York: ACM, 2011 : 1517 -1520.
  • 10李静,林鸿飞,李瑞敏.基于情感向量空间模型的歌曲情感标签预测模型[J].中文信息学报,2012,26(6):45-50. 被引量:7

二级参考文献210

共引文献124

同被引文献178

引证文献11

二级引证文献48

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部