基于关联挖掘的图情类主要期刊参考文献分析
|
摘要
【目的/意义】鉴于参考文献在期刊论文发表过程中的重要性,提出基于关联规则的期刊论文参考文献分析研究,探究图书情报学领域学科结构现状及参考文献选择路径中的潜在知识信息。【方法/过程】以图情学领域八种主要期刊为研究对象,获取其2012-2016年刊发的全部论文作为样本,结构化处理数据后运用关联规则挖掘方法,从参考文献作者相关性分析、主题分析和来源期刊相关性分析三个方面展开研究,探究其参考文献引用过程中隐藏的关联规则并加以横向对比分析。【结果/结论】通过研究发现:参考文献作者同被引现象很大程度上受到合著关系的影响,但同被引相关性与合作强度不成正比,合作关系普遍存在小团体现象,且不具对称性;参考文献标题中提取的特征关键词与目标期刊论文关键词之间存在一定的相关关系,同时通过对图情类期刊的热点关键词分析可得,国内图书馆学和情报学虽然融合趋势明显,但研究领域并没有很好地结合,研究对象和主题也不尽相同;来源期刊之间存在较强的相关关系,会随着时间阶段产生一定的变化。
【Purpose/significance】Given the importance of references in the publishing process of journal papers, an analysis and research of the references based on association rules is proposed to explore the potential knowledge information and its portability in the reference selection path.【Method/process】This paper takes papers published in eight kinds of Library and Information influential journals during 2012 and 2016 as the research object. Processing unstructured text data to be structured and then using association rules method of data mining, the research develops mainly from correlation analysis of the references' authors, theme and source journals correlation analysis three parts, to explore the implied correlation between journal references and horizontal comparative analysis.【Result/conclusion】The study found that: (1)Source papers' authors co-cited phenomenon is largely affected by cooperative relationships, but the correlation is not directly proportional to the intensity of cooperation, clique phenomenon is widespread in cooperative relations, and is not symmetrical. (2) The words extracted from references' title are related to papers' keywords. At the same time, through the analysis of hot keywords of the Library and Information influential journals, we found that domestic Library Science and Information Science are relatively independent, their research fields are not well mixed, and research object and subject are different. (3) There is a strong correlation between the source journals, and it will change with time phase. The research results help authors understand research status of Library and Information Science and provide some decision-making suggestions for the author to read and write, and for editorial department to deal with the process of the reviewing and publication of papers and the recommendation, reading and citation of references.
【Purpose/significance】Given the importance of references in the publishing process of journal papers, an analysis and research of the references based on association rules is proposed to explore the potential knowledge information and its portability in the reference selection path.【Method/process】This paper takes papers published in eight kinds of Library and Information influential journals during 2012 and 2016 as the research object. Processing unstructured text data to be structured and then using association rules method of data mining, the research develops mainly from correlation analysis of the references' authors, theme and source journals correlation analysis three parts, to explore the implied correlation between journal references and horizontal comparative analysis.【Result/conclusion】The study found that: (1)Source papers' authors co-cited phenomenon is largely affected by cooperative relationships, but the correlation is not directly proportional to the intensity of cooperation, clique phenomenon is widespread in cooperative relations, and is not symmetrical. (2) The words extracted from references' title are related to papers' keywords. At the same time, through the analysis of hot keywords of the Library and Information influential journals, we found that domestic Library Science and Information Science are relatively independent, their research fields are not well mixed, and research object and subject are different. (3) There is a strong correlation between the source journals, and it will change with time phase. The research results help authors understand research status of Library and Information Science and provide some decision-making suggestions for the author to read and write, and for editorial department to deal with the process of the reviewing and publication of papers and the recommendation, reading and citation of references.
引文
1 王丹丹.科学数据规范引用关键问题探析[J].图书情报工作, 2015, 59(8):42-47.
2 陈丹,刘应竹.基于引文的参考文献评价功能的实现及其正确途径[J].编辑学报, 2014, 26(2):107-109.
3 胡冉,邵敏.基于引文分析法的外文图书引进工作研究——以清华大学图书馆为例[J].图书情报工作,2013, 57(21):71-75.
4 张娅,龚新刚,李秀萍,等.档案引文分析在馆藏期刊评价中的应用[J].图书情报工作, 2013, 57(S2):192-195.
5 张金柱,韩涛,王小梅.利用参考文献的学科分类分析图书情报领域的学科交叉性[J].图书情报工作, 2013, 57(1):108-111.
6 Dejian Yu, Zeshui Xu, Witold Pedrycz, et al. Information sciences 1968–2016:A retrospective analysis with text mining and bibliometric[J]. Information Sciences, 2017,(9), 418-419.
7 孙建辉,张荣丽.图书馆的文献服务对研究生参考文献研究行为的影响[J].图书情报工作,2015,59(S1):166-169.
8 李海林,叶益,林春培,等.基于关联规则的期刊论文参考文献决策分析研究——以《控制与决策》期刊为例[J].情报科学, 2018,(8):84-89,100.
9 李牧南.技术预见研究热点的演进分析:内容挖掘视角[J].科研管理, 2018,39(3):141-153.
10 丁洁兰,刘细文,杨立英,等.科学计量方法在科技政策研究中应用的实证研究[J].图书情报工作, 2017, 61(24):77-86.
11 李霞,王连喜,路美秀,等.基于复合词生成的网络热点话题识别及描述算法[J].图书情报工作, 2016, 60(23):128-134
12 阮光册,夏磊.基于词共现关系的检索结果知识关联研究[J].情报学报, 2017, 36(12):1247-1254.
13 王玲,李树林,徐培培,等.基于频繁项集树的时态关联规则挖掘算法[J].控制与决策, 2018, 33(4):591-599.
14 岳俊举,冯立杰,冯奕程,等.基于多维技术创新地图与关联规则挖掘的技术机会识别方法研究[J].情报学报,2017, 36(8):798-808.
15 刘雪立,方红玲,周志新,等.我国自引率不同的科技期刊文献计量学特征的对照研究[J].编辑学报, 2011, 23(1):8-11.
16 杨勇,达庆利.谁能获得国家自然科学基金的资助?[J].管理工程学报, 2017, 31(3):52-58.
17 赵蓉英,许丽敏.文献计量学发展演进与研究前沿的知识图谱探析[J].中国图书馆学报, 2010, 36(5):60-68.
2 陈丹,刘应竹.基于引文的参考文献评价功能的实现及其正确途径[J].编辑学报, 2014, 26(2):107-109.
3 胡冉,邵敏.基于引文分析法的外文图书引进工作研究——以清华大学图书馆为例[J].图书情报工作,2013, 57(21):71-75.
4 张娅,龚新刚,李秀萍,等.档案引文分析在馆藏期刊评价中的应用[J].图书情报工作, 2013, 57(S2):192-195.
5 张金柱,韩涛,王小梅.利用参考文献的学科分类分析图书情报领域的学科交叉性[J].图书情报工作, 2013, 57(1):108-111.
6 Dejian Yu, Zeshui Xu, Witold Pedrycz, et al. Information sciences 1968–2016:A retrospective analysis with text mining and bibliometric[J]. Information Sciences, 2017,(9), 418-419.
7 孙建辉,张荣丽.图书馆的文献服务对研究生参考文献研究行为的影响[J].图书情报工作,2015,59(S1):166-169.
8 李海林,叶益,林春培,等.基于关联规则的期刊论文参考文献决策分析研究——以《控制与决策》期刊为例[J].情报科学, 2018,(8):84-89,100.
9 李牧南.技术预见研究热点的演进分析:内容挖掘视角[J].科研管理, 2018,39(3):141-153.
10 丁洁兰,刘细文,杨立英,等.科学计量方法在科技政策研究中应用的实证研究[J].图书情报工作, 2017, 61(24):77-86.
11 李霞,王连喜,路美秀,等.基于复合词生成的网络热点话题识别及描述算法[J].图书情报工作, 2016, 60(23):128-134
12 阮光册,夏磊.基于词共现关系的检索结果知识关联研究[J].情报学报, 2017, 36(12):1247-1254.
13 王玲,李树林,徐培培,等.基于频繁项集树的时态关联规则挖掘算法[J].控制与决策, 2018, 33(4):591-599.
14 岳俊举,冯立杰,冯奕程,等.基于多维技术创新地图与关联规则挖掘的技术机会识别方法研究[J].情报学报,2017, 36(8):798-808.
15 刘雪立,方红玲,周志新,等.我国自引率不同的科技期刊文献计量学特征的对照研究[J].编辑学报, 2011, 23(1):8-11.
16 杨勇,达庆利.谁能获得国家自然科学基金的资助?[J].管理工程学报, 2017, 31(3):52-58.
17 赵蓉英,许丽敏.文献计量学发展演进与研究前沿的知识图谱探析[J].中国图书馆学报, 2010, 36(5):60-68.