基于VSM的美国一流大学图书馆网站导航文本调查与分析  

Analyzing Website Navigation Features of Top U.S. Academic Libraries

作　　者：尹相权[1] 李书宁[1] Yin Xiangquan Li Shuning(Beijing Normal University Library, Beijing 100875, China)

机构地区：[1]北京师范大学图书馆,北京100875

出　　处：《数据分析与知识发现》2017年第3期90-95,共6页Data Analysis and Knowledge Discovery

摘　　要：【目的】通过分析美国一流大学图书馆网站导航文本的特点,为国内大学图书馆导航建设提供建议。【方法】结合一流大学应具有一流学科、一流教师和一流学生的思路选取选取15所美国一流大学图书馆,基于标签云和文本挖掘模型VSM,分析导航文本词维度上的共性和特异性,并结合《2016年美国图书馆状况报告》进行数据验证。【结果】与人工调研相比,基于VSM模型的统计方法可以更直观、快速地给出基础调研结果,调研结果可供进一步深度文本分析参考。【局限】仅选取一级导航、二级导航和首页标题板块概况导航文本。【结论】基于文本数据挖掘模型的统计方法可以更直观、快速地给出基础调研结果,为高校图书馆网站导航建设提供参考。[Objective] This paper studies the navigation features of top academic library websites from the United States, aiming to improve the services of their Chinese counterparts. [Methods] First, we identified library websites of the top 15 U.S. universities and downloaded their navigation texts. Second, we analyzed the similarities and differences among these texts with tag cloud and Vector Space Model. Finally, we examined our findings with the ＂2016 State of America＇s Libraries Report＂. [Results] The proposed method was intuitive and generated analysis results fast, which could be further processed with text mining techniques. [Limitations] Only retrieved the first and, second levels of navigation as well as titles of the homepages. [Conclusions] The proposed model provides useful information for the academic libraries in China.

关 键 词：美国大学图书馆 网站导航 向量空间模型 调查分析 

分 类 号：G250.7[文化科学—图书馆学]