融合WK和CSK Co-word Model的共词分析法

发布时间：2018-03-10 21:36

本文选题：的差别　切入点：提出共现频率计算时应对主要主题词进行加权计算(WKWA)。唐晓波和肖璐[]指出自标引的关键词不能全面描述论文的主题内容　出处：《电子科技》2017年02期 　论文类型：期刊论文

【摘要】：基于关键词的共词分析方法是利用关键词在文献中出现的频次高低来确定该领域研究热点的方法。传统的基于关键词的共词分析方法只是简单的统计关键词出现的绝对次数,忽略了关键词以及文献的内在特性以及关键词在概念上的重复性,造成结果的不准确性。文中提出了一种融合关键词加权模型(WK Co-word Model)和同义关键词合并模型(CSK Co-word Model)的共词分析法,该方法根据关键词自身的特征以及关键词所在载体文献的特征对关键词进行加权处理,同时以同义词词林为基础,计算关键词之间的词语相似度,合并同义关键词。该方法既强调了关键词之间权重的不同,又消除了同义词对结果准确性造成的影响。仿真实验表明,该方法提高了共词分析的准确性。
[Abstract]:The coterm analysis method based on keywords is a method to determine the research hotspot in this field by using the frequency of keywords appearing in the literature. The traditional cowords analysis method based on keywords is only a simple statistical method of absolute number of keywords appearing. Ignoring the key words and the intrinsic characteristics of the literature, and the conceptual repeatability of the keywords, In this paper, we present a coterm analysis method which combines the weighted keyword model (WK Co-word Model) and the synonymous keyword combination model (CSK Co-word Model). In this method, the keywords are weighted according to their own characteristics and the characteristics of the documents where the keywords are located. At the same time, the similarity between the keywords is calculated on the basis of the synonym forest. The method not only emphasizes the weight difference among keywords, but also eliminates the influence of synonyms on the accuracy of the results. The simulation results show that the method improves the accuracy of coterm analysis.
【作者单位】：上海理工大学光电信息与计算机工程学院;
【基金】：国家自然科学基金资助项目(61170277) 上海市教委科研创新基金资助项目(02120557)
【分类号】：TP391.1;G252.7

【相似文献】