基于CTM模型的学科间知识交流分析——以图书情报学和计算机信息科学间的知识交流为例
|
摘要
[目的/意义]为展现学科间知识交流的研究内容、探测学科间知识交流的趋势,利用CTM模型从文献中提取学科间知识交流的主题,研究对促进学科知识发展与创新、推动学科间的知识交流与借鉴具有重要意义。[方法/过程]以图书情报学(ISLS)和计算机信息科学(CIS)为例,利用"二八定律"选取两学科核心期刊,以其中2008—2016年的文献数据为研究对象,借助开源软件R,利用困惑度和似然函数值变化拐点来确定模型主题数目,并根据文档—主题分布计算各个主题的主题强度,进而分析主题变化趋势。[结果/结论]该方法能够较全面地展现学科间知识交流的主题。分析主题—词汇分布,发现两学科在知识交流中主题内容不断丰富、交流主题更加细化。
[Purpose/significance] This paper aims to show the content of knowledge exchange between disciplines and explore the trend of interdisciplinary knowledge exchange. Using CTM,the paper extracts the themes of knowledge exchange between disciplines from literatures,which is of great significance to promote the development and innovation of subject knowledge,as well as the knowledge exchange and reference. [Method/process]Taking the Information Science & Library Science( ISLS) and Computer Information System( CIS) as examples,the paper applies the Pareto principle to select the core journals of the two disciplines,and uses the articles of the selected core journals from 2008 to 2016 as data. With the help of open-source software R,the number of model themes is determined by the inflection point of the perplexity and Log-likelihood function values,and the subject intensity of each theme is calculated according to the document-theme distribution. On this basis,the paper further analyzes the development trend of themes. [Result/conclusion] This method can show the themes of inter-disciplinary knowledge exchange comprehensively. Through analyzing the theme-vocabulary distribution,it is found that the content gets enriched and the themes become increasingly detailed in the knowledge exchange of the two disciplines.
[Purpose/significance] This paper aims to show the content of knowledge exchange between disciplines and explore the trend of interdisciplinary knowledge exchange. Using CTM,the paper extracts the themes of knowledge exchange between disciplines from literatures,which is of great significance to promote the development and innovation of subject knowledge,as well as the knowledge exchange and reference. [Method/process]Taking the Information Science & Library Science( ISLS) and Computer Information System( CIS) as examples,the paper applies the Pareto principle to select the core journals of the two disciplines,and uses the articles of the selected core journals from 2008 to 2016 as data. With the help of open-source software R,the number of model themes is determined by the inflection point of the perplexity and Log-likelihood function values,and the subject intensity of each theme is calculated according to the document-theme distribution. On this basis,the paper further analyzes the development trend of themes. [Result/conclusion] This method can show the themes of inter-disciplinary knowledge exchange comprehensively. Through analyzing the theme-vocabulary distribution,it is found that the content gets enriched and the themes become increasingly detailed in the knowledge exchange of the two disciplines.
引文
[1]杨瑞仙.知识交流内涵和类型探讨[J].情报理论与实践,2014,37(3):12-15.
[2]邱均平,杨思洛,宋艳辉.知识交流研究现状可视化分析[J].中国图书馆学报,2012,38(2):78-89.
[3]彭继东,谭宗颖.纳米科技学科领域的知识交流——基于期刊引文网络的分析[J].图书情报工作,2011,55(4):15-18.
[4]KISS I Z,BROOM M,CRAZE P G,et al.Can epidemic models describe the diffusion of topics across disciplines?[J].Journal of Informetrics,2010,4(1):74-82.
[5]YAN E.Finding knowledge paths among scientific disciplines[J].Journal of the Association for Information Science&Technology,2013,65(11):2331-2347.
[6]COWAN R,JONARD N.Network structure and the diffusion of knowledge[J].Research Memorandum,2004,28(8):1557-1575.
[7]XIAO Y,LU L Y Y,LIU J S,et al.Knowledge diffusion path analysis of data quality literature:a main path analysis[J].Journal of Informetrics,2014,8(3):594-605.
[8]HUR W.The patterns of knowledge spillovers across technology sectors evidencedin patent citation networks[J].Scientometrics,2017,111(2):595-619.
[9]邱均平,王菲菲.基于博客社区好友链接的知识交流状况分析——以科学网博客为例[J].图书情报知识,2011(6):25-33.
[10]王惠,王树乔.图书情报类期刊知识交流效率评价及影响因素研究[J].情报科学,2017(3):134-138.
[11]王旻霞,赵丙军.中国图书情报学跨学科知识交流特征研究——基于CCD数据库的分析[J].情报理论与实践,2015,38(5):94-99.
[12]马秀峰,张莉,李秀霞.我国图书情报学与新闻传播学间的学科知识交流与融合分析[J].情报杂志,2017,36(2):59-65.
[13]张艺蔓,马秀峰,程结晶.融合引文内容和全文本引文分析的知识流动研究[J].情报杂志,2015(11):50-54.
[14]关智远,陈仕吉.跨学科知识交流研究综述[J].情报杂志,2016,35(3):153-158.
[15]王曰芬,傅柱,陈必坤.采用LDA主题模型的国内知识流研究结构探讨:以学科分类主题抽取为视角[J].现代图书情报技术,2016,32(4):8-19.
[16]曹娟,张勇东,李锦涛,等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10):1780-1787.
[17]王燕霞,邓伟.CTM与SVM相结合的文本分类方法[J].计算机工程,2010,36(22):203-205.
[18]BLEI D M,LAFFERTY J D.Correction:a correlated topic model of Science[J].Annals of Applied Statistics,2007,1(1):17-35.
[19]关鹏,王曰芬.科技情报分析中LDA主题模型最优主题数确定方法研究[J].现代图书情报技术,2016,32(9):42-50.
[20]马长林,杨正良,谢罗迪.文本分类中CTM模型的优化和可视化应用研究[J].计算机工程与科学,2017,39(3):599-604.
[21]GRN B,HORNIK K,LEEUW J D,et al.Topicmodels:an R package for fitting topic models[J].Journal of Statistical Software,2011,40(13):2011.
[22]杨海霞,高宝俊,孙含林.基于LDA挖掘计算机科学文献的研究主题[J].现代图书情报技术,2016,32(11):20-26.
[23]肖建国.试论文本挖掘及其应用[J].图书馆学研究,2008(4):22-24.
[24]李广建,黄崑.用户模型及其学习方法[J].现代图书情报技术,2002,18(6):24-27.
[25]邱均平,吕红.近五年国际图书情报学研究热点、前沿及其知识基础——基于17种外文期刊知识图谱的可视化分析[J].图书情报知识,2013(3):4-15.
[26]邱均平,曹洁.不同学科间知识扩散规律研究——以图书情报学为例[J].情报理论与实践,2012,35(10):1-5.
[2]邱均平,杨思洛,宋艳辉.知识交流研究现状可视化分析[J].中国图书馆学报,2012,38(2):78-89.
[3]彭继东,谭宗颖.纳米科技学科领域的知识交流——基于期刊引文网络的分析[J].图书情报工作,2011,55(4):15-18.
[4]KISS I Z,BROOM M,CRAZE P G,et al.Can epidemic models describe the diffusion of topics across disciplines?[J].Journal of Informetrics,2010,4(1):74-82.
[5]YAN E.Finding knowledge paths among scientific disciplines[J].Journal of the Association for Information Science&Technology,2013,65(11):2331-2347.
[6]COWAN R,JONARD N.Network structure and the diffusion of knowledge[J].Research Memorandum,2004,28(8):1557-1575.
[7]XIAO Y,LU L Y Y,LIU J S,et al.Knowledge diffusion path analysis of data quality literature:a main path analysis[J].Journal of Informetrics,2014,8(3):594-605.
[8]HUR W.The patterns of knowledge spillovers across technology sectors evidencedin patent citation networks[J].Scientometrics,2017,111(2):595-619.
[9]邱均平,王菲菲.基于博客社区好友链接的知识交流状况分析——以科学网博客为例[J].图书情报知识,2011(6):25-33.
[10]王惠,王树乔.图书情报类期刊知识交流效率评价及影响因素研究[J].情报科学,2017(3):134-138.
[11]王旻霞,赵丙军.中国图书情报学跨学科知识交流特征研究——基于CCD数据库的分析[J].情报理论与实践,2015,38(5):94-99.
[12]马秀峰,张莉,李秀霞.我国图书情报学与新闻传播学间的学科知识交流与融合分析[J].情报杂志,2017,36(2):59-65.
[13]张艺蔓,马秀峰,程结晶.融合引文内容和全文本引文分析的知识流动研究[J].情报杂志,2015(11):50-54.
[14]关智远,陈仕吉.跨学科知识交流研究综述[J].情报杂志,2016,35(3):153-158.
[15]王曰芬,傅柱,陈必坤.采用LDA主题模型的国内知识流研究结构探讨:以学科分类主题抽取为视角[J].现代图书情报技术,2016,32(4):8-19.
[16]曹娟,张勇东,李锦涛,等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10):1780-1787.
[17]王燕霞,邓伟.CTM与SVM相结合的文本分类方法[J].计算机工程,2010,36(22):203-205.
[18]BLEI D M,LAFFERTY J D.Correction:a correlated topic model of Science[J].Annals of Applied Statistics,2007,1(1):17-35.
[19]关鹏,王曰芬.科技情报分析中LDA主题模型最优主题数确定方法研究[J].现代图书情报技术,2016,32(9):42-50.
[20]马长林,杨正良,谢罗迪.文本分类中CTM模型的优化和可视化应用研究[J].计算机工程与科学,2017,39(3):599-604.
[21]GRN B,HORNIK K,LEEUW J D,et al.Topicmodels:an R package for fitting topic models[J].Journal of Statistical Software,2011,40(13):2011.
[22]杨海霞,高宝俊,孙含林.基于LDA挖掘计算机科学文献的研究主题[J].现代图书情报技术,2016,32(11):20-26.
[23]肖建国.试论文本挖掘及其应用[J].图书馆学研究,2008(4):22-24.
[24]李广建,黄崑.用户模型及其学习方法[J].现代图书情报技术,2002,18(6):24-27.
[25]邱均平,吕红.近五年国际图书情报学研究热点、前沿及其知识基础——基于17种外文期刊知识图谱的可视化分析[J].图书情报知识,2013(3):4-15.
[26]邱均平,曹洁.不同学科间知识扩散规律研究——以图书情报学为例[J].情报理论与实践,2012,35(10):1-5.