您的位置:首页  > 论文页面

数字图书馆中的热点研究主题提取

发表时间:2009-10-31  浏览量:1535  下载量:692
全部作者: 韩晓晖,马军
作者单位: 山东大学计算机科学与技术学院
摘 要: 首先提出一种结构化论文相似度计算方法,然后提出一种从数字图书馆的论文集中提取一定时间段内热点研究主题的方法,其主要思想为:基于提出的结构化相似度度量将论文集合聚类成簇,每个簇被看作一个主题;其后,计算出每个主题的热度,有最高热度值的主题被选择为热点主题。实验结果证明了结构化论文相似度计算方法和热点研究主题提取方法的有效性。
关 键 词: 计算机应用;主题检测;相似度计算;数字图书馆;科技论文
Title: Hot research topic extraction in digital libraries
Author: HAN Xiaohui, MA Jun
Organization: School of Computer Science and Technology, Shandong University
Abstract: This paper proposes a structural paper similarity calculation method, and then proposes a novel approach to extract hot research topics during a time period from the corpus of digital libraries. A structural paper similarity measure was presented firstly. Then the similarity measure was used to group the papers into topics. Finally, the hotness of each topic was evaluated and the topics with highest hotness were selected as hot research topics. The results show the effectiveness of the paper similarity calculation method and hot research topic extraction method.
Key words: computer application; topic detection; similarity calculation; digital libraries; scientific paper
发表期数: 2009年10月第20期
引用格式: 韩晓晖,马军. 数字图书馆中的热点研究主题提取[J]. 中国科技论文在线精品论文,2009,2(20):2143-2149.
 
0 评论数 0
暂无评论
友情链接