文章摘要
刘奕杉,王玉琳,李明鑫.词频分析法中高频词阈值界定方法适用性的实证分析[J].数字图书馆论坛,2017,(9):42~49
词频分析法中高频词阈值界定方法适用性的实证分析
An Empirical Analysis for the Applicability of the Methods of Definition of High-Frequency Words in Word Frequency Analysis
  
DOI:
中文关键词: 高频词;文献计量学;词频分析
英文关键词: High-Frequency Word;Bibliometrics;Word Frequency Analysis
基金项目:
作者单位
刘奕杉 东北师范大学 
王玉琳 东北师范大学 
李明鑫 东北师范大学 
摘要点击次数: 2060
全文下载次数: 2301
中文摘要:
      词频分析法是文献计量学的重要分析方法之一,而确定高频词阈值是进行词频分析的必要前提,高频词阈值的选取不仅决定词频分析法的分析结果,而且对整个分析研究都有着极其重要的影响.本文首先以近三年国内运用词频分析法展开研究的文献为调研基础,发现目前学界常用的高频词阈值选取方法主要有自定义选取法、高低频词界定公式选取法、普赖斯公式选取法及混合选取法四类;其次,以个人知识管理领域的文献为研究对象,对前三类高频词阈值选取方法分别进行取值计算并做领域热点聚类分析,对比验证聚类结果,同时以此结果为基础讨论高频词阈值选择对分析结果的影响及其合理性;最后,指出我国学界在高频词阈值选取方面存在主观性强、方法原理不明、改进方法适用性不明,高低频词界定公式和普赖斯公式适用性尚待研究等问题.
英文摘要:
      Word frequency analysis method is one of the important analysis methods in bibliometrics, and the selection of high-frequency word is a necessary premise. It is to say that the selection of high-frequency word determines the results of word frequency analysis, impacts the whole analysis program. First, the paper cleared up the nearly three years papers in China by using word frequency analysis method for hot spots analysis, and found four common classes selections of high-frequency word methods mainly include:the author set the selection method, Donohue's formula selection, price formula selection and mixed selection. Secondly, we use the literature of personal knowledge management as the research object, and calculate the frond three kinds of high frequency words selections respectively, and compare the results with clustering analysis, then we discuss the effect and applicability of high-frequency word threshold selection on the analysis results. At last, the paper pointed out that there were some problems, such as the subjective is high, principle is unclear, improved methods' principle is unclear, the Donohue's formula and price formula's applicability are stil unsure, in the study of high-frequency word threshold selection in our country.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮