文献题录信息挖掘方法及其软件SATI的实现.pdfVIP

如您使用SATI 软件进行研究，还请将如下一文列入参考文献： 刘启元,叶鹰.文献题录信息挖掘方法及其软件SATI 的实现[J],信息资源管理学报,2012,(1):50-58. 文献题录信息挖掘方法及其软件SATI 的实现 ——以中外图书情报学为例 刘启元 叶 鹰 （浙江大学信息资源管理系，杭州310027） [摘 要] 利用C#编程技术基于.NET 平台设计开发出文献题录信息统计分析工具软件 SATI，可导入处理EndNote 格式、NoteExpress 格式及NoteFirst 格式的国内文献题录数据和 HTML 格式的WoS 国际文献题录数据，进行数据格式的转换、字段信息的抽取、词条频次 的统计和知识单元共现矩阵、词条频率逐年分布矩阵及文档词条矩阵的构建，以辅助生成聚 类图、多维尺度图谱、网络知识图谱、策略坐标图等可视化结果。以2006—2010 年中外图 书情报学各十种具有代表性的核心期刊刊载的17446 篇论文数据为实例，基于聚类与多维尺 度分析结果，呈现出中外图书情报学三大主要研究领域，并结合共词分析与社会网络分析方 法，通过绘制共现网络知识图谱与策略坐标图，进一步揭示研究领域结构的内部联系及其特 征。 [关键词] SATI 共词分析 聚类分析 多维尺度分析 知识图谱 策略坐标图 A Study on Mining Bibliographic Records by Designed Software SATI: Case Study on Library and Information Science Liu Qiyuan Ye Ying (Department of Information Resource Management, Zhejiang University, Hangzhou 310027) [Abstract] A bibliographic information analysis software named SATI (Statistical Analysis Toolkit for Informetrics) is developed using C# based on Microsoft .NET platform. The national data fitting EndNote, NoteExpress and NoteFirst can be imported into SATI as well as international data as HTML (output from WoS). For the purpose of getting the clustering graph, multidimensional scaling map, network knowledge map, and strategic diagram, four basic functions including transforming raw data into XML, extracting selected elements, counting terms ’ frequency and building knowledge units ’ matrices have been implemented. Taking 17444 articles published in ten core Library and Information Science journals both at home and abroad during 2006 to 2010 as the sample, this paper revealed three potential research fields of LIS research area based on the consistency between clustering analysis and multidimensional scaling analysis results, and figured out relations and features of subject areas by interpreting the network knowledge map and str