词频分析法中高频词阈值界定方法适用性的实证分析

发布时间：2018-09-10 17:41

【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。
[Abstract]:Word frequency analysis is one of the important analytical methods in bibliometrics, and the determination of the threshold of high-frequency words is a necessary prerequisite for word frequency analysis. The selection of the threshold of high-frequency words not only determines the analysis results of word frequency analysis. Moreover, it has an extremely important influence on the whole analysis and research. First of all, based on the literature about the use of word frequency analysis in recent three years, this paper finds out that the methods of selecting the threshold of high-frequency words are mainly self-defined method and high-low frequency word definition formula selection method, which are commonly used in academic circles at present. Secondly, taking the literature in the field of personal knowledge management as the research object, the first three kinds of high-frequency word threshold selection methods are calculated, and the hot spot clustering analysis is done for the first three kinds of high-frequency word threshold selection methods. At the same time, the influence and rationality of the threshold selection of high-frequency words on the analysis results are discussed. Finally, it is pointed out that there is a strong subjectivity in the selection of high-frequency word threshold in Chinese academic circles, and the principle of the method is not clear. The applicability of the improved method is unknown, and the applicability of the high and low frequency word definition formula and Price formula remains to be studied.
【作者单位】：东北师范大学信息科学与技术学院;
【分类号】：G353.1

【相似文献】