摘要
本文利用自组织映射(SOM)人工神经网络方法对学术期刊按其主题进行可视化组织.在修改常见的SOM显示方式统一距离矩阵(U-matrix)的基础上提出增强型U-matrix及新的SOM显示方式'属性方差矩阵'(AV-matrix),构造了'关键属性投影'方法,以53种有代表性的图书情报类英文期刊为例,将期刊按其主题分为19个类,识别各类期刊之间的关键差异主题,并分析各类期刊在关键差异主题上的特点.
This paper aimed to visually organize the academic journals according to their subjects with the Self- Organizing Map (SOM) technique. An Enhanced Unified Distance Matrix (U-matrix) was presented with some modifications on U-matrix, a conventional SOM display. A novel SOM display named Attribute Variance Matrix (AV-matrix) was also proposed and a method of Key Attribute Projection was constructed. Fifty-three typical English journals in the field of Library and Information Science (LIS) were selected as experimental samples and categorized into nineteen clusters. The subjects which contributed the most to the differences among journals were identified and the characteristics of journal clusters in terms of subjects were analyzed.
出处
《情报学报》
CSSCI
北大核心
2011年第2期183-191,共9页
Journal of the China Society for Scientific and Technical Information
基金
中国博士后科学基金面上资助项目,国家自然科学基金青年科学基金,武汉大学自主科研项目(人文社会科学),'中央高校基本科研业务费专项资金'资助
关键词
自组织映射
期刊
主题
可视化
self-organizing map
journal
subject
visualization