摘要
关于OPAC数据的研究多基于一所或几所高校的少量数据,从单个用户角度展开,忽视了用户群体的阅读偏好。文章采集我国103所高校图书馆"热门图书"数据,从用户群体阅读偏好角度进行聚类分析:提取部分"热门图书"作为反映高校用户群体阅读偏好的特征,为每所高校生成用户群体阅读偏好向量,对这些高校进行AP (Affinity Propagation)聚类。对聚类结果进行分析发现:部分类簇高校的图书类别分布、图书题名高频词、高频学科主题词存在明显差异。
Current researches on OPAC data are mostly based on a small amount of data from one or several university libraries,paying attention to single library users while ignoring the group reading preference of library users.On the basis of "hot books" data collected from 103 Chinese university libraries,a cluster analysis is made of library users’ group reading preference.It first selects some hot books as key characteristics to reflect the group reading preference of library users,then works out the group reading preference vector for each university,and finally conducts an AP(Affinity Propagation) clustering for these universities.The clustering result shows that there are significant differences in the distribution of book categories,the high-frequency words in book titles and the high-frequency subject words.
作者
张恒
章成志
周清清
ZHANG Heng;ZHANG Chengzhi;ZHOU Qingqing
出处
《图书馆论坛》
CSSCI
北大核心
2020年第5期132-141,共10页
Library Tribune
基金
中央高校基本科研业务费专项资金资助项目“国际顶尖人才资源信息与分布研究”(项目编号:30917013133)
2018年江苏省研究生科研与实践创新计划项目“学术图书的内容抽取与知识关联研究”(项目编号:SJCX18_0136)研究成果。
关键词
热门图书
阅读偏好
群体阅读偏好
高校图书馆
hotbook
reading preference
group reading preference
university library
