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。
Co-word cluster analysis based on keyword is a method for determining the research focus point in a field by the times keywords appear in the literature. Traditional co-word cluster analysis based keyword method simply calculates the absolute times keywords appear, ignoring the inherent characteristics of keywords and literature as well as the repeated keywords in concept, thus inaccurate results. This paper proposes a new co-word cluster analysis method that merges the keyword weighted model （WK Co-word Model） with synonymous keywords combined model （CSK Co-word Model）. The method weights the keyword according to characteristics of keywords and the literature that keywords appear, and calculates the similarity of the keywords and combined synonymous keywords, which not only emphasize the different weight between the keywords, but also eliminates the effects caused by synonyms key- words. Experiment shows that the new method improves the accuracy of co-word cluster analysis.
出处
《电子科技》
2017年第2期110-113,118,共5页
Electronic Science and Technology
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)
关键词
关键词
加权
共词分析
同义词
相似度
keyword
weighted
co-word analysis
synonyms
similarity
