基于社会网络关注度的学科前沿热点挖掘
|
摘要
从科研文献数据中挖掘出学科前沿热点是目前学术界和工业界亟待解决的问题.社会网络可以及时反映信息传播的实际受欢迎程度,故提出一种基于社会网络关注度的学科前沿热点挖掘方法.首先通过数据相关性分析、相关属性划分以及社会网络关注度因子挖掘,构建文献热度评价模型.同时,采用文档主题生成模型(latent dirichlet allocation,LDA)从文献热度评价模型挖掘的科研文献中识别出该学科的前沿热点.最后,在"artificial intelligence and image processing"学科的数据集上构建评价模型并进行多组对比实验,结果表明提出的方法有效提高了学科热点挖掘结果的前沿性,热点主题在时间维度上更具时效性.
Mining subject hotspots from scientific literature data was an urgent problem in the academic and industry field. The existing methods couldn't consider the popularity of a scientific paper as a propagating object in the social network,which reflected the practical popularity of information diffusion timely. A hotspot mining method was proposed. Firstly,the literature popularity evaluation model was constructed for social network attention scoring. The model consisted of data correlation analyzing,correlated attributes partitioning and social network attention degree factor mining. Meanwhile,LDA model was adopted to discover the hotspots from the scientific papers. Finally,contrastive experiments were performed on the topic of artificial intelligence and image processing. The results showed that the method could timely discover recent hotspots,so it was effective in hot frontiers mining.
Mining subject hotspots from scientific literature data was an urgent problem in the academic and industry field. The existing methods couldn't consider the popularity of a scientific paper as a propagating object in the social network,which reflected the practical popularity of information diffusion timely. A hotspot mining method was proposed. Firstly,the literature popularity evaluation model was constructed for social network attention scoring. The model consisted of data correlation analyzing,correlated attributes partitioning and social network attention degree factor mining. Meanwhile,LDA model was adopted to discover the hotspots from the scientific papers. Finally,contrastive experiments were performed on the topic of artificial intelligence and image processing. The results showed that the method could timely discover recent hotspots,so it was effective in hot frontiers mining.
引文
[1]王慧.社会网络分析在学科热点分析中的实证研究[D].镇江:江苏大学,2010.
[2]LUO Y,ZHAO S L,LI X C,et al.Text keyword extraction method based on word frequency statistics[J].Journal of computer applications,2016,36(3):718-725.
[3]XIE P.Study of international anticancer research trends via co-word and document co-citation visualization analysis[M].New York:Springer-Verlag,2015.
[4]XIE Y F.Hotspots of ecological and environmental risk research in China based on multidimensional scaling analysis and cluster analysis[J].Advanced materials research,2013,807(2):641-646.
[5]XIE Y F.Hotspots and trend of water transfer research in China based on clustered analysis and social network analysis[J].International journal of digital content technology and its applications,2013,7(3):534-541.
[6]江贺,陈信,张静宣,等.软件仓库挖掘领域:贡献者和研究热点[J].计算机研究与发展,2016,52(12):2768-2782.
[7]XU T Y,QU H N,ZHAO S S,et al.The visualization analysis of research hotspot and frontier technology of the smart power distribution and utilization based on the cite space[J].Energy and power engineering,2017,9(4):515-524.
[8]ZHAO R,WEI M.Impact evaluation of open source software:an Altmetrics perspective[J].Scientometrics,2017,110(2):1-17.
[9]YU D,LI D F,MERIGO J M,et al.Mapping development of linguistic decision making studies[J].Journal of intelligent and fuzzy systems,2016,30(5):2727-2736.
[10]PENG Y L,LIN A W,WANG K,et al.Global trends in DEM-related research from 1994 to 2013:a bibliometric analysis[J].Scientometrics,2015,105(1):347-366.
[11]YIN Z Y,CHEN D Y,LI B F.Global regulatory T-cell research from 2000 to 2015:a bibliometric analysis[J].Plos one,2016,11(9):0162099.
[12]WU Y,DUAN Z G.Visualization analysis of author collaborations in schizophrenia research[J].BMC psychiatry,2015,15(1):1-8.
[13]徐佳俊,杨飏,姚天昉,等.基于LDA模型的论坛热点话题识别和追踪[J].中文信息学报,2016,30(1):43-49.
[14]DING W Y,CHEN C M.Dynamic topic detection and tracking:a comparison of HDP,C-word,and cocitation methods[J].Journal of the association for information science and technology,2014,65(10):2084-2097.
[15]吴岘辉,张晖,杨春明,等.一种话题相关的微博意见领袖挖掘算法[J].小型微型计算机系统,2014,35(10):2296-2301.
[16]由庆斌,韦博,汤珊红.基于补充计量学的论文影响力评价模型构建[J].图书情报工作,2014,58(22):5-11.
[17]HOLSAPPLE C W,JOHNSON L E,MANAKYAN H et al.Business computing research journals:a normalized citation analysis[J].Journal of management information systems,2015,11(1):131-140.
[2]LUO Y,ZHAO S L,LI X C,et al.Text keyword extraction method based on word frequency statistics[J].Journal of computer applications,2016,36(3):718-725.
[3]XIE P.Study of international anticancer research trends via co-word and document co-citation visualization analysis[M].New York:Springer-Verlag,2015.
[4]XIE Y F.Hotspots of ecological and environmental risk research in China based on multidimensional scaling analysis and cluster analysis[J].Advanced materials research,2013,807(2):641-646.
[5]XIE Y F.Hotspots and trend of water transfer research in China based on clustered analysis and social network analysis[J].International journal of digital content technology and its applications,2013,7(3):534-541.
[6]江贺,陈信,张静宣,等.软件仓库挖掘领域:贡献者和研究热点[J].计算机研究与发展,2016,52(12):2768-2782.
[7]XU T Y,QU H N,ZHAO S S,et al.The visualization analysis of research hotspot and frontier technology of the smart power distribution and utilization based on the cite space[J].Energy and power engineering,2017,9(4):515-524.
[8]ZHAO R,WEI M.Impact evaluation of open source software:an Altmetrics perspective[J].Scientometrics,2017,110(2):1-17.
[9]YU D,LI D F,MERIGO J M,et al.Mapping development of linguistic decision making studies[J].Journal of intelligent and fuzzy systems,2016,30(5):2727-2736.
[10]PENG Y L,LIN A W,WANG K,et al.Global trends in DEM-related research from 1994 to 2013:a bibliometric analysis[J].Scientometrics,2015,105(1):347-366.
[11]YIN Z Y,CHEN D Y,LI B F.Global regulatory T-cell research from 2000 to 2015:a bibliometric analysis[J].Plos one,2016,11(9):0162099.
[12]WU Y,DUAN Z G.Visualization analysis of author collaborations in schizophrenia research[J].BMC psychiatry,2015,15(1):1-8.
[13]徐佳俊,杨飏,姚天昉,等.基于LDA模型的论坛热点话题识别和追踪[J].中文信息学报,2016,30(1):43-49.
[14]DING W Y,CHEN C M.Dynamic topic detection and tracking:a comparison of HDP,C-word,and cocitation methods[J].Journal of the association for information science and technology,2014,65(10):2084-2097.
[15]吴岘辉,张晖,杨春明,等.一种话题相关的微博意见领袖挖掘算法[J].小型微型计算机系统,2014,35(10):2296-2301.
[16]由庆斌,韦博,汤珊红.基于补充计量学的论文影响力评价模型构建[J].图书情报工作,2014,58(22):5-11.
[17]HOLSAPPLE C W,JOHNSON L E,MANAKYAN H et al.Business computing research journals:a normalized citation analysis[J].Journal of management information systems,2015,11(1):131-140.