摘要
文献计量学中的共被引分析是一种基于参考文献的共现关系而建立共被引网络的分析方法,可用来展现引文的结构和聚类关系。将文献计量学中的共被引思想拓展到对互联网用户音乐收藏行为的分析中,进行音乐分类,是一种新的尝试。通过从某在线音乐网站随机选择500名网络用户作为研究样本,从他们的音乐收藏夹中提取歌手的共被引次数矩阵,计算Jaccard系数矩阵,进一步进行共被引分析,利用社会网络分析工具绘制歌手共被引网络。可以将这种方法进一步拓展到其他互联网文化消费领域,改进互联网文化消费的可视化搜索,挖掘互联网用户的消费兴趣,增强用户体验。
The co-citation analysis in bibliometrics is a method to display the structure and classification of the citation references based on their cooccurrence relations.Songs/singers co-collection relationship by online music web users is referred from the co-citation analysis theory in bibliometrics field.Based on this relationship,the authors get the singers co-collection network,which can serve as a new approach to classify singers.In this paper,the authors select 500 web users randomly from one online music website as research samples,extract the singers co-collection matrix from the users' music favorites,then make co-collection analysis and network analysis when Jaccard matrix calculated,This method could also be expanded into other Internet culture consumption field,and will be helpful for improving the visualization search technologe,digging the the interests of Internet user,and eventually enhancing the Internet user experience.
出处
《情报学报》
CSSCI
北大核心
2011年第5期471-478,共8页
Journal of the China Society for Scientific and Technical Information
基金
教育部高等学校博士学科点专项科研基金(批准号20090041110001)
国家自然科学基金项目(批准号70773015)
关键词
音乐分类
音乐可视化搜索
共被引
网络计量
music classification
music visualization search
co-citation
webmetrics