期刊学术水平综合评价低维逐次投影寻踪建模与实证研究
|
摘要
[目的/意义]对综合性人文、社会科学类期刊学术水平进行科学、合理的评价,为高校、科研院所、期刊学界和业界提供决策依据。[方法/过程]采用文献综述法,以《中国学术期刊影响因子年报》的指标为基础,构建由总被引频次等17个指标组成的期刊评价指标体系。对198种综合性人文、社会科学类期刊的样本数据进行正态化和规格化预处理,并建立四维逐次投影寻踪模型,再采用矢量合成法构建低维逐次投影寻踪模型。[结果/结论]实现对期刊学术水平、各个评价指标重要性等排序和分类,《中国社会科学》显著优于其他期刊,其次是《浙江大学学报》;他引总引比指标最重要,其次是平均引文数,排名前10位指标占总权重的78%;提出了提升期刊学术水平的措施和建议;拓展了期刊评价的新方法;论述了期刊综合评价常用方法 (模型)的优缺点和适用性等。
[Purpose/significance]This paper aims to scientifically and reasonably evaluate the academic levels of the humanity and social science periodicals(HSSPs),and to provide the decision-making references for colleges,research institutions,periodicals,and the academic industry.[Method/process]According to the literature metrology,and based on the indicators used by the "Annual Report on Impact Factors of Chinese Academic Journals"(ARIFCAJ),the comprehensive evaluation indicators' system(CEIS) for the academic level of HSSPs is established in the paper.The system is composed of 17 indicators including the total cited frequencies et al.the sample data of the 198 HSSPs is positively normalized.The four-dimension sequential projection pursuit clustering(SPPC) model is established,and the low-dimensional SPPC(LDSPPC) model is built up by the vector syntheses.[Result/conclusion]The rank and classification of the academic level of the HSSPs and the evaluation indicators are carried out according to the established LDSPPC model.The study results show that the academic level of the"Journal of Social Sciences in China"(JSSC) is the highest,and the"Journal of Zhejiang University"follows up.Among the 17 indicators,the total cites ratio on other cites is the most important,and the second is the average cited number.The most important ten indicators take up 78% of the total weight.According to the rank of the indicators,the measures and suggestions to improve the academic level of HSSPs are put forward.The LDSPPC model is a new method to evaluate the academic level of HSSPs,and the merits,defects and application of other common evaluation methods or models are also discussed.
[Purpose/significance]This paper aims to scientifically and reasonably evaluate the academic levels of the humanity and social science periodicals(HSSPs),and to provide the decision-making references for colleges,research institutions,periodicals,and the academic industry.[Method/process]According to the literature metrology,and based on the indicators used by the "Annual Report on Impact Factors of Chinese Academic Journals"(ARIFCAJ),the comprehensive evaluation indicators' system(CEIS) for the academic level of HSSPs is established in the paper.The system is composed of 17 indicators including the total cited frequencies et al.the sample data of the 198 HSSPs is positively normalized.The four-dimension sequential projection pursuit clustering(SPPC) model is established,and the low-dimensional SPPC(LDSPPC) model is built up by the vector syntheses.[Result/conclusion]The rank and classification of the academic level of the HSSPs and the evaluation indicators are carried out according to the established LDSPPC model.The study results show that the academic level of the"Journal of Social Sciences in China"(JSSC) is the highest,and the"Journal of Zhejiang University"follows up.Among the 17 indicators,the total cites ratio on other cites is the most important,and the second is the average cited number.The most important ten indicators take up 78% of the total weight.According to the rank of the indicators,the measures and suggestions to improve the academic level of HSSPs are put forward.The LDSPPC model is a new method to evaluate the academic level of HSSPs,and the merits,defects and application of other common evaluation methods or models are also discussed.
引文
[1]熊国经,熊玲玲,陈小山.组合评价和复合评价模型在学术期刊评价优越性的实证研究[J].现代情报,2017,37(1):81-88.
[2]曾伟,田时中,田家华.科技期刊学术影响力综合评价模型与实证[J].中国科技期刊研究,2016,27(3):316-323.
[3]叶继元.学术期刊质量评价具有多元性与复杂性[J].清华大学学报:哲学社会科学版,2015,30(2):182-187.
[4]虞玉华,楼文高.体育类期刊学术水平综合评价与实证研究---基于决策者偏好的投影寻踪建模技术[J].北京体育大学学报,2015,38(12):46-54.
[5]俞立平,刘骏.主成分分析与因子分析法适合科技评价吗?---以学术期刊评价为例[J].现代情报,2018,38(6):73-79.
[6]陈小山,陈国福,张瑞.基于因子分析和SEM模型的期刊评价指标结构关系研究[J].情报科学,2016,34(10):61-64.
[7]熊国经,熊玲玲,陈小山.基于因子分析与TOPSIS法在学术期刊评价中的改进研究[J].情报杂志,2016,35(7):196-200.
[8]张和平,陈齐海.基于PLSR的学术期刊影响力影响因素研究[J].情报科学,2018,36(2):53-57.
[9]张发明,刘志平.组合评价方法研究综述[J].系统工程学报,2017,32(4):557-569.
[10]FRIEDMAN J H,TUKEY J W.A projection pursuit algorithm for exploratory data analysis[J].IEEE Transactions on Computers,1974,23(9):881-890.
[11]楼文高,干瑞娟,李坦.基于投影寻踪模型的图书馆成效(绩效)评估研究[J].图书情报工作,2017,61(9):65-73.
[12]楼文高,熊聘,冯国珍,于晓虹.影响投影寻踪聚类建模的关键因素分析与实证研究[J].数理统计与管理,2017,36(5):783-801.
[13]楼文高,乔龙.投影寻踪聚类建模理论的新探索与实证研究[J].数理统计与管理,2015,34(1):47-58.
[14]于晓虹,楼文高.线上供应链金融风险评价的低维逐次投影寻踪建模研究[J].金融理论与实践,2017(12):49-52.
[15]于晓虹,楼文高,冯国珍,等.低维逐次投影寻踪聚类模型综合评价方法、装置及应用[EB/OL].[2017-12-01].http://www.patentstar.cn.
[16]于晓虹,楼文高,康海燕.供应链企业信贷风险动态聚类投影寻踪建模与实证研究[J].数学的实践与认识,2018,48(11):31-40.
[17]张雯雰,刘华艳.改进的群搜索优化算法在MATLAB中的实现[J].电脑与信息技术,2010,18(3):44-46.
[18]俞立平,武夷山.学术期刊客观赋权评价新方法---指标难度赋权法[J].现代图书情报技术,2011(4):64-70.
[19]刘雪立.论期刊影响因子人为操纵的识别[J].编辑学报,2018,30(1):98-101.
[20]邓履翔,王维朗,陈灿华.欺诈引用---一种新的不当引用行为[J].中国科技期刊研究,2018,29(3):237-241.
[21]任胜利,李家林,金碧辉,张玉华.我国部分科技期刊参考文献和被引用情况统计分析[J].编辑学报,2001,13(5):261-263.
[22]TABACHNICK B G,FIDELL L S.Using multivariate statistics[M].6thEdition.New Jersey:Pearson Education,Inc.,2013.
[23]王先甲,汪磊.基于马氏距离的改进型TOPSIS在供应商选择中的应用[J].控制与决策,2012,27(10):1566-1570.
[24]ZHANG G,PATUWO E,HU M.Forecasting with artificial neural networks:the state of the art[J].International Journal of Forecasting,1998,14(1):55-56.
[25]STATSOFT.Electronic statistics textbook[EB/OL].[2015-09-08].http://www.statsoft.com/textbook.
[26]楼文高.活性污泥工艺神经网络模拟与控制研究[D].上海:同济大学,2005.
[27]楼文高.基于人工神经网络的三江平原土壤质量综合评价与预测模型[J].中国管理科学,2002,10(1):79-83.
[2]曾伟,田时中,田家华.科技期刊学术影响力综合评价模型与实证[J].中国科技期刊研究,2016,27(3):316-323.
[3]叶继元.学术期刊质量评价具有多元性与复杂性[J].清华大学学报:哲学社会科学版,2015,30(2):182-187.
[4]虞玉华,楼文高.体育类期刊学术水平综合评价与实证研究---基于决策者偏好的投影寻踪建模技术[J].北京体育大学学报,2015,38(12):46-54.
[5]俞立平,刘骏.主成分分析与因子分析法适合科技评价吗?---以学术期刊评价为例[J].现代情报,2018,38(6):73-79.
[6]陈小山,陈国福,张瑞.基于因子分析和SEM模型的期刊评价指标结构关系研究[J].情报科学,2016,34(10):61-64.
[7]熊国经,熊玲玲,陈小山.基于因子分析与TOPSIS法在学术期刊评价中的改进研究[J].情报杂志,2016,35(7):196-200.
[8]张和平,陈齐海.基于PLSR的学术期刊影响力影响因素研究[J].情报科学,2018,36(2):53-57.
[9]张发明,刘志平.组合评价方法研究综述[J].系统工程学报,2017,32(4):557-569.
[10]FRIEDMAN J H,TUKEY J W.A projection pursuit algorithm for exploratory data analysis[J].IEEE Transactions on Computers,1974,23(9):881-890.
[11]楼文高,干瑞娟,李坦.基于投影寻踪模型的图书馆成效(绩效)评估研究[J].图书情报工作,2017,61(9):65-73.
[12]楼文高,熊聘,冯国珍,于晓虹.影响投影寻踪聚类建模的关键因素分析与实证研究[J].数理统计与管理,2017,36(5):783-801.
[13]楼文高,乔龙.投影寻踪聚类建模理论的新探索与实证研究[J].数理统计与管理,2015,34(1):47-58.
[14]于晓虹,楼文高.线上供应链金融风险评价的低维逐次投影寻踪建模研究[J].金融理论与实践,2017(12):49-52.
[15]于晓虹,楼文高,冯国珍,等.低维逐次投影寻踪聚类模型综合评价方法、装置及应用[EB/OL].[2017-12-01].http://www.patentstar.cn.
[16]于晓虹,楼文高,康海燕.供应链企业信贷风险动态聚类投影寻踪建模与实证研究[J].数学的实践与认识,2018,48(11):31-40.
[17]张雯雰,刘华艳.改进的群搜索优化算法在MATLAB中的实现[J].电脑与信息技术,2010,18(3):44-46.
[18]俞立平,武夷山.学术期刊客观赋权评价新方法---指标难度赋权法[J].现代图书情报技术,2011(4):64-70.
[19]刘雪立.论期刊影响因子人为操纵的识别[J].编辑学报,2018,30(1):98-101.
[20]邓履翔,王维朗,陈灿华.欺诈引用---一种新的不当引用行为[J].中国科技期刊研究,2018,29(3):237-241.
[21]任胜利,李家林,金碧辉,张玉华.我国部分科技期刊参考文献和被引用情况统计分析[J].编辑学报,2001,13(5):261-263.
[22]TABACHNICK B G,FIDELL L S.Using multivariate statistics[M].6thEdition.New Jersey:Pearson Education,Inc.,2013.
[23]王先甲,汪磊.基于马氏距离的改进型TOPSIS在供应商选择中的应用[J].控制与决策,2012,27(10):1566-1570.
[24]ZHANG G,PATUWO E,HU M.Forecasting with artificial neural networks:the state of the art[J].International Journal of Forecasting,1998,14(1):55-56.
[25]STATSOFT.Electronic statistics textbook[EB/OL].[2015-09-08].http://www.statsoft.com/textbook.
[26]楼文高.活性污泥工艺神经网络模拟与控制研究[D].上海:同济大学,2005.
[27]楼文高.基于人工神经网络的三江平原土壤质量综合评价与预测模型[J].中国管理科学,2002,10(1):79-83.