[Purpose/Significance] Academic discourse power is an integral part of China’s international discourse power system and the main manifestation of national political, economic, scientific and technological soft power. Analyzing the evaluation methods of academic discourse power and comprehensively comparing the advantages, disadvantages and stability of different methods are helpful to provide certain references for the evaluation of academic discourse power. [Method/Process] Six assignment methods without comprehensive evaluation values were used for single model evaluation, and fuzzy combination evaluation was carried out on the results after passing the non-parametric test to reduce the tendency of single evaluation and improve the credibility of evaluation, and BP neural network was introduced to construct neural network model based on gradient descent algorithm. [Result/Conclusion] This paper constructs an academic discourse power evaluation system based on three dimensions: academic leadership based on the innovation leading index, academic influence based on the citation analysis index and academic communication capacity based on the Altmetrics index. Fuzzy Borda evaluation allows for a combination of high and low evaluation scores and evaluation ranking for a single model, and realizes the internal combination of objective information, which has higher accuracy than the single evaluation model. Based on this, an evaluation model of academic discourse power integrated with BP neural network was constructed.
Zhao Rongying
,
Zhu Weijie
,
Zhang Zhaoyang
,
Li Xinlai
. Research on the Evaluation Method of Academic Discourse Power by Fusing on BP Neural Network[J]. Library and Information Service, 2022
, 66(11)
: 50
-58
.
DOI: 10.13266/j.issn.0252-3116.2022.11.006
[1] 王旭.中国学术期刊话语权评价理论框架建构研究[J].图书情报工作,2021,65(12):83-92.
[2] 邵娅芬.经济学科的国际学术话语权研究[D].上海:上海交通大学,2011.
[3] 王兴.国际学术话语权视角下的大学学科评价研究——以化学学科世界1387所大学为例[J].清华大学教育研究,2015,36(3):64-75.
[4] 余波,赵蓉英,邱均平.中国学者学术话语权评价指标体系的构建与应用研究——以国际生物学为例[J].情报科学,2021,39(12):105-112,125.
[5] 吴坚,贺蕾,盘劲呈,等.体育学学者学术合作网络对学术话语权影响的实证研究[J].武汉体育学院学报,2022,56(1):28-35.
[6] 闫奕文,张海涛,孙思阳,等.基于BP神经网络的政务微信公众号信息传播效果评价研究[J].图书情报工作,2017,61(20):53-62.
[7] 俞立平.基于神经网络的非线性学术评价方法选择研究[J].情报理论与实践,2021,44(1):63-70,56.
[8] 郑杭生.学术话语权与中国社会学发展[J].中国社会科学,2011(2):27-34,4,220.
[9] 赵蓉英,张晓曦,刘卓著,等.话语权与话语权评价解析[J].情报理论与实践,2021,44(11):15-22.
[10] 赵蓉英,王旭,余波,等.中国话语权的评价科学研究框架构建[J].图书与情报,2019(4):122-131.
[11] 权衡.中国经济学:话语权,范式转换及其他[J].探索与争鸣,2006(3):38-40
[12] 王旭,赵蓉英.融合Altmetrics的话语引导力指标与被引频次关系研究——以中国英文学术期刊为例[J].情报理论与实践,2021,44(7):37-43,36.
[13] 赵蓉英,魏绪秋,王建品.引文分析研究与进展[J].情报学进展,2018,12:50-80.
[14] 杨岳,王贤文.Altmetrics十年发展综述[J].情报杂志,2021,40(11):136-146.
[15] 科睿唯安.2020研究前沿[EB/OL].[2021-06-27].https://solutions.clarivate.com.cn/blog/20201113/.
[16] 王春柳,杨永辉,邓霏,等.文本相似度计算方法研究综述[J].情报科学,2019,37(3):158-168.
[17] 于鹏.逻辑公式间的Jaccard距离及其应用[J].计算机科学与探索,2020,14(11):1975-1980.
[18] SUN J. A note on principal component analysis for multi-dimensional data[J].Statistics & probability letter,2000(46):69-73.
[19] LEIGH S. A user's guide to principal components[J].Technometrics,1993,35(1):83-85.
[20] CHEN M, LU D, ZHA L. The comprehensive evaluation of China's urbanization and effects on resources and environment[J].Journal of geographical sciences,2010,20(1):17-30.
[21] GUPTA R, TRIPATHI R, MICHALEK J, et al. An exact test for the mean of a normal distribution with a known coefficient of variation[J].Computational statistics & data analysis,1985, 3:219-226.
[22] 王应明.运用离差最大化方法进行多指标决策与排序[J].系统工程与电子技术,1998(7):26-28,33.
[23] 谭学瑞,邓聚龙.灰色关联分析:多因素统计分析新方法[J].统计研究, 1995(3):46-48.
[24] 徐林明,林志炳,李美娟,等.基于模糊Borda法的动态组合评价方法及其应用研究[J].中国管理科学,2017,25(2):165-173.
[25] 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):73-79,137.
[26] FABRIGAR L R, WEGENER D T, MACCALLUM R C, et al. Evaluating the use of exploratory factor analysis in psychological research[J]. Psychological methods,1999, 4(3):272-299.
[27] 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.作者贡献说明:赵蓉英:论文选题、框架设计、论文指导;朱伟杰:数据分析、论文撰写、论文修改;张兆阳:数据获取、数据处理;李新来:数据分析、论文修改。
