一种基于组块分析的共现词提取方法  

A Co-word Abstraction Method Based on Chunk Parsing

作　　者：赵国荣[1,2] 王文剑[1] 杨光[2] 

机构地区：[1]山西大学计算机与信息技术学院,山西太原030006 [2]山西大学图书馆,山西太原030006

出　　处：《情报科学》2017年第12期129-135,共7页Information Science

基　　金：国家自然科学基金(61673249);山西省科技基础条件平台建设项目(2014091001-0101)

摘　　要：【目的/意义】文献计量学方法是研究学科发展趋势、捕捉学科前沿热点的一种定量化的方法。共词分析是一种重要的文献计量学方法,一般将作者选定的关键词作为最常用的词源。但科技论文中作者给出的关键词个数有限,会存在缺失或者不能充分表达主题等情况,从而导致丢失一些重要的共现关系。【方法/过程】本文采用组块分析的方法从文章标题中提取短语或词作为作者给定关键词的有益补充。【结果/结论】以中文句法分析领域的文献作为研究对象进行实验,结果证明增补后的关键词列表增加了共现关系,优化了聚类结果。最后对中文句法分析领域发展趋势及研究热点进行了分析。[Purpose/significance] Bibliometric is a quantitative method to study the development trend and catch the hot spot of the subject. Co-word analysis is an important method of bibliometric. The keywords that the author selected are in- dexing and convenience, so they have become the most commonly used word source for co-word analysis. But the amount of keywords given by the author in the science and technology papers is limited, it will cause a lack of keywords, or the key- words are not accurate enough that cannot fully express the theme, etc. This situation will lead to the loss of some important co-existing relationships among the keywords. [Method/process] In this paper, chunk parsing method was used to extract chunks from the title of the article, these phrases or words are useful supplement to the author＇s keywords. [Result/conclu- sion] The literature in Chinese parsing field was used as the research object. Experimental results show that the supple- mented keywords list increases the co-occurrence relations among the keywords, and it makes clusters more closely. Final- ly, strategic diagram and social network analysis are used to catch the hotspot and trend in Chinese parsing field.

关 键 词：组块分析 共词网络 战略图 社会网络分析 中文句法分析 

分 类 号：G254[文化科学—图书馆学]