一种基于爬山算法的学科领域热点探索模型
|
摘要
[目的/意义]热点发掘一直是图书馆领域的重点研究内容之一,通过共引和共词分析探测学科领域热点都存在一定的局限性,针对存在的问题,借鉴爬山法进行改进研究。为适应大数据时代的热点挖掘问题,将计算机算法与文献关键词分析相结合,构建研究热点分析模型,探索分析研究热点的方法,对于学科领域研究热点具有重要意义。[方法/过程]采用改进的Apriori-Hill-Mount爬山算法,拓展关键词共现分析和聚类方法,构建了篇与篇之间的关键词的关联规则挖掘模型;通过实证研究科学计量学及图书情报学两个领域,论证该模型的可行性。[结果/结论]通过对上述两个领域的文献进行篇-篇关键词强关联规则挖掘和分析,分别得到各自领域的的研究热点;其结果与采用Citespace及Spss呈现的热点有很高的吻合度。但该算法揭示的信息更全面、更直观地呈现研究主题热点,对学科热点探索体系方法及手段具有一定的补充及辅助作用。该方法是在一个局部范围内进行搜索,且阈值的设定也具有人为因素,因此得到的可能是局部的优化,其更大范围的可适性有待进一步论证。
[Purpose/Significance]Hotspot mining has always been one of the key research contents in the field of library. There are some limitations in detecting hotspots through co-citation and Co-word analysis. In view of the existing problems, the mountain climbing method is used in order to improve the research analysis. To solve the hotspot mining problems in the era of big data, it is of great significance to combine computer algorithms with keyword analysis in literature, to construct a hotspot analysis model and to explore new methods of hotspot analysis.[Method/Process]By adopting the improved Apriori-Hill-Mount climbing algorithm, expanding keyword co-occurrence analysis and clustering method, an association rule mining model of the keywords between the texts was constructed. The feasibility of the model was verified through the empirical study of the two subjects of Bibliometrics and Library and Information Science. [Result/Conclusion]Through the mining and analysis of the strong association rules of keywords in the above two fields, the research hotspots were obtained respectively. The results have a high coincidence and similarity with the hot spots or clusters presented by Citespace and Spss, but the information revealed by this model is more comprehensive and more intuitive. It is supplementary and helpful to the research methods and means of discipline hotspots.The method is to search within a local range, and the threshold setting also contains human factors, so the result obtained may be a local optimization. The greater scope of its suitability needs to be further explored.
[Purpose/Significance]Hotspot mining has always been one of the key research contents in the field of library. There are some limitations in detecting hotspots through co-citation and Co-word analysis. In view of the existing problems, the mountain climbing method is used in order to improve the research analysis. To solve the hotspot mining problems in the era of big data, it is of great significance to combine computer algorithms with keyword analysis in literature, to construct a hotspot analysis model and to explore new methods of hotspot analysis.[Method/Process]By adopting the improved Apriori-Hill-Mount climbing algorithm, expanding keyword co-occurrence analysis and clustering method, an association rule mining model of the keywords between the texts was constructed. The feasibility of the model was verified through the empirical study of the two subjects of Bibliometrics and Library and Information Science. [Result/Conclusion]Through the mining and analysis of the strong association rules of keywords in the above two fields, the research hotspots were obtained respectively. The results have a high coincidence and similarity with the hot spots or clusters presented by Citespace and Spss, but the information revealed by this model is more comprehensive and more intuitive. It is supplementary and helpful to the research methods and means of discipline hotspots.The method is to search within a local range, and the threshold setting also contains human factors, so the result obtained may be a local optimization. The greater scope of its suitability needs to be further explored.
引文
[1] 鑫鑫,魏凤. 基于CiteSpace的国内竞争情报领域研发态势及特点分析[J]. 知识管理论坛,2017,2(2):128-135.
[2] 刘则渊,陈悦,朱晓宇. 普赖斯对科学学理论的贡献——纪念科学计量学之父普赖斯逝世30周年[J]. 科学学研究,2013,31(12):1761-1772.
[3] 钟伟金.李佳.共词分析法研究(一)—共词分析的过程与方式[J].情报杂志,2008,27( 5) : 70-72.
[4] 陈悦,陈超美,刘则渊,等. CiteSpace知识图谱的方法论功能[J]. 科学学研究,2015,33(2):242-253.
[5] 陈定权,朱维凤. 关联规则与图书馆书目推荐[J]. 情报理论与实践,2009,32(6):81-84.
[6] 邓春林,邹凯. 基于关联规则的图书馆读者兴趣度实证分析[J]. 情报理论与实践,2009,32(4):93-95,108.
[7] 李勇男,梅建明. 先验原理在涉恐情报分析中的应用研究[J]. 情报杂志,2017,36(8):23-26.
[8] 岳韶华,周国安,张纳温,等. 基于灰色关联分析的组网情报处理性能评价[J]. 空军工程大学学报(自然科学版),2009,10(1):69-73.
[9] 何超,张玉峰. 基于语义关联分析的商务情报分析算法研究[J]. 情报杂志,2013,32(4):134-137
[10] 许海云,方曙.基于专利功效矩阵的技术主题关联分析及核心专利挖掘[J].情报学报,2014,33(2):158-166.
[11] 雷雪,侯人华,曾建勋. 关联规则在领域知识推荐中的应用研究[J]. 情报理论与实践,2014,37(12):67-70,66.
[12] 胡雪,封化民,李明伟等. 数据挖掘中一种增强的Apriori算法分析[J]. 信息网络安全,2015(11):77-83.
[13] 鄢铁平,孙燕,廖炜等. 关联规则支持下坡面土壤侵蚀评价指标体系构建——以鄂西北丹江库区郧西县为例[J]. 中国水利,2011(22):49-51,55.
[14] Tsamardinos I, Brown L E, Aliferis C F. The max-min hill-climbing bayesian network structure learning algorithm[J]. Machine Learning, 2006, 65(1):31-78.
[15] 邹金花. 关联规则在图书馆中的应用[J]. 现代情报,2013,33(5):17-20.
[16] 周友行,何清华,邓伯禄.一种改进的爬山法优化求解冗余机械手运动学逆解[J].机器人,2003(1):35-38.
[17] 龙永文. 基于遗传算法的永磁无刷直流电机优化设计研究[D].南京:东南大学,2005.
[18] 张健,王蔚.基于支持度与置信度阈值优化技术的关联分类算法[J].计算机应用,2007(12):3032-3034,3038.
[19] 刘星沙,谭利球,熊拥军. 关联规则挖掘算法及其应用研究[J]. 计算机工程与科学,2007(1):83-85,108.
[20] 侯剑华,杨秀财,李莲姬.科学计量学研究中心转移特征的可视化分析[J].科学学研究,2018,36(2):211-220.
[21] 谢萍,郑莎. 美国图书情报学专业TOP高校2006-2011年研究热点聚类分析[J]. 图书情报研究,2013,6(3):38-43,32.
[22] 胡浪. 2012-2016年国内图书情报学研究现状及热点探析——基于13种中文图情核心期刊的计量分析[J]. 图书馆工作与研究,2018(2):67-76.
[2] 刘则渊,陈悦,朱晓宇. 普赖斯对科学学理论的贡献——纪念科学计量学之父普赖斯逝世30周年[J]. 科学学研究,2013,31(12):1761-1772.
[3] 钟伟金.李佳.共词分析法研究(一)—共词分析的过程与方式[J].情报杂志,2008,27( 5) : 70-72.
[4] 陈悦,陈超美,刘则渊,等. CiteSpace知识图谱的方法论功能[J]. 科学学研究,2015,33(2):242-253.
[5] 陈定权,朱维凤. 关联规则与图书馆书目推荐[J]. 情报理论与实践,2009,32(6):81-84.
[6] 邓春林,邹凯. 基于关联规则的图书馆读者兴趣度实证分析[J]. 情报理论与实践,2009,32(4):93-95,108.
[7] 李勇男,梅建明. 先验原理在涉恐情报分析中的应用研究[J]. 情报杂志,2017,36(8):23-26.
[8] 岳韶华,周国安,张纳温,等. 基于灰色关联分析的组网情报处理性能评价[J]. 空军工程大学学报(自然科学版),2009,10(1):69-73.
[9] 何超,张玉峰. 基于语义关联分析的商务情报分析算法研究[J]. 情报杂志,2013,32(4):134-137
[10] 许海云,方曙.基于专利功效矩阵的技术主题关联分析及核心专利挖掘[J].情报学报,2014,33(2):158-166.
[11] 雷雪,侯人华,曾建勋. 关联规则在领域知识推荐中的应用研究[J]. 情报理论与实践,2014,37(12):67-70,66.
[12] 胡雪,封化民,李明伟等. 数据挖掘中一种增强的Apriori算法分析[J]. 信息网络安全,2015(11):77-83.
[13] 鄢铁平,孙燕,廖炜等. 关联规则支持下坡面土壤侵蚀评价指标体系构建——以鄂西北丹江库区郧西县为例[J]. 中国水利,2011(22):49-51,55.
[14] Tsamardinos I, Brown L E, Aliferis C F. The max-min hill-climbing bayesian network structure learning algorithm[J]. Machine Learning, 2006, 65(1):31-78.
[15] 邹金花. 关联规则在图书馆中的应用[J]. 现代情报,2013,33(5):17-20.
[16] 周友行,何清华,邓伯禄.一种改进的爬山法优化求解冗余机械手运动学逆解[J].机器人,2003(1):35-38.
[17] 龙永文. 基于遗传算法的永磁无刷直流电机优化设计研究[D].南京:东南大学,2005.
[18] 张健,王蔚.基于支持度与置信度阈值优化技术的关联分类算法[J].计算机应用,2007(12):3032-3034,3038.
[19] 刘星沙,谭利球,熊拥军. 关联规则挖掘算法及其应用研究[J]. 计算机工程与科学,2007(1):83-85,108.
[20] 侯剑华,杨秀财,李莲姬.科学计量学研究中心转移特征的可视化分析[J].科学学研究,2018,36(2):211-220.
[21] 谢萍,郑莎. 美国图书情报学专业TOP高校2006-2011年研究热点聚类分析[J]. 图书情报研究,2013,6(3):38-43,32.
[22] 胡浪. 2012-2016年国内图书情报学研究现状及热点探析——基于13种中文图情核心期刊的计量分析[J]. 图书馆工作与研究,2018(2):67-76.