题名与关键词在文献内容揭示中的对比研究——基于农产品品牌评价领域  

Contrastive Research on Titles and Keywords Represented Main Ideal of Documents——Based on the Field of Agricultural Brand Evaluation

作　　者：张紫玄 王雪颖[1,2] 王昊[1,2] ZHANG Zi-xuan WANG Xue-ying WANG Hao(School of lnformation Management, Nanjing University, Nanjing 210023, China Jiangsu Key Laboratory of Data Engineering and Knowledge Service, Nanjing University, Nanjing 210023, China)

机构地区：[1]南京大学信息管理学院,江苏南京210023 [2]南京大学江苏省数据工程与知识服务重点实验室,江苏南京210023

出　　处：《情报科学》2017年第10期88-93,171,共7页Information Science

基　　金：国家社会科学基金项目(12BTQ037);国家自然科学基金青年项目(71503121)

摘　　要：【目的/意义】随着数字文本数量的增加,题名与关键词作为文本表示方式之一,对文本的揭示作用日益突显。【方法/过程】基于我国农产品品牌评价领域的相关文献,对文章的题名与关键词分别进行了聚类分析,并对两者的聚类结果进行了对比,发现基于题名与基于关键词的聚类结果不完全一致,但存在交集,进而将题名与关键词叠加后进行聚类,经过分析认为聚类结果得到了改善。【结果/结论】发现题名与关键词对文献内容的揭示程度不同,且均不能良好地概括农产品品牌评价领域的文献内容,二者融合后结果得到改善;同时对今后研究中论文题名与关键词的选取提出了一些建议。【Purpose/significance】As the number of digital text increasing, the title and keyword begin play a more important role in text representation.【Method/process】In this paper, clustering method will be taken to find the difference of their ability and the gap between this two in revealing literature content. This study based on the papers on area of in China＇s agricultural product brand evaluation in CNKI and carried out K-means clustering between the title and keywords respectively. Then the clustering analysis of the integration of title and keyword shows that both the title-based clustering and the keyword-based clustering can not provide a good overview of the focus of current research on brand evaluation of agricultural products and there＇re differences between them.【Result/conclusion】After the clustering analysis of the integration of title and keyword, the result has been improved. The conclusion is there is a gap between the title and keywords in the field of agricultural product brand evaluation and both of them can not reveal the content of academic papers well. The fusion of them can improve the effect. Some suggestions were given on the selection of the title and keyword in the future research.

关 键 词：文本聚类 K-MEANS 内容揭示 

分 类 号：G254.9[文化科学—图书馆学]