融合链接分析和内容分析视角的主题门户网站信息组织研究——以国际组织WHO为例

摘要

门户网站的信息组织方式是图情领域研究的重要内容之一,而面向特定领域人群、特定用户需求的主题门户则更具有研究意义.本文综合应用了链接分析、文本相似度计算、社会网络分析、聚类分析、块模型分析等方法,提出融合链接分析与内容分析为视角,对主题门户网站的信息组织方式进行研究,并选取世界卫生组织(WHO)网站为实验案例,提出了该主题门户信息组织的优化方案,以期为已有的主题门户网站的改进提供参考.实例证明,通过一定的方式,有选择性的增加内容相似度高的主题间链接,可以有效实现主题门户网站的信息组织优化.研究也表明链接关系是一种隐含的语义关系,网站信息组织不能仅考虑语义相似度高的页面,也要考虑语义相似度低但存在链接关系的页面.%The method of information organization on the portal website is one of the important concerns of library and information science, and it is important to explore the theme portal of a specific area of the crowd and specific user needs. By integrating hyperlink analysis, text similarity calculation, social network analysis, cluster analysis, model analysis and so on, this paper examines the method of information organization on the theme portal website from the perspectives of link analysis and content analysis. Furthermore, this paper selects WHO as an experimental case and proposes the optimization of the theme portal website. Results show that, through a certain method, a selec-tive increase of content similarity among the theme of the link can effectively achieve the optimization of information organization on the theme portal website. The results also show that the link relationship is a kind of implicit seman-tic relation. Information organization of websites can not only consider the semantic similarity of the page, but also consider low semantic similarity and high related links pages.