Please wait a minute...
Advanced Search
数据分析与知识发现  2018, Vol. 2 Issue (6): 48-57     https://doi.org/10.11925/infotech.2096-3467.2017.1124
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
自然权重对非线性科技评价的影响及纠正研究*——以TOPSIS方法评价为例
俞立平1,3, 宋夏云2, 王作功3()
1浙江工商大学管理工程与电子商务学院 杭州 310018
2浙江财经大学会计学院 杭州 310018
3贵州财经大学金融学院 贵阳 550025
Impacts and Corrections of Natural Weight on Nonlinear Sci-tech Reviews——Case Study of TOPSIS Method
Yu Liping1,3, Song Xiayun2, Wang Zuogong3()
1School of Management and e-Business, Zhejiang Gongshang University, Hangzhou 310018, China
2School of Accounting, Zhejiang University of Finance and Economics, Hangzhou 310018, China
3Finance School, Guizhou University of Finance and Economics, Guiyang 550025, China
全文: PDF (619 KB)   HTML ( 2
输出: BibTeX | EndNote (RIS)      
摘要 

目的】本文提出隐含在科技评价指标中的数据自然权重问题, 并提出了修正方法。【方法】以JCR2016数学期刊和TOPSIS评价方法为例, 分析自然权重对非线性评价方法的影响, 提出动态最大均值逼近标准化方法, 以消除自然权重的影响。【结果】自然权重对非线性评价方法影响较大, 对于加权类非线性评价方法, 设计权重、自然权重和评价方法共同影响实际权重, 对于非加权类线性评价方法, 自然权重和评价方法影响实际权重; 自然权重消除后可以有效降低评价方法对实际权重的影响, 从而充分发挥设计权重的作用, 这符合评价公理; 指标数据分布特点也会影响实际权重。【局限】用来消除自然权重的动态最大均值逼近标准化方法是一种逼近算法, 均值标准化结果难以完全相等。【结论】在科技评价中必须重视自然权重问题, 这是一种系统误差, 消除后才能保证评价公平。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
俞立平
宋夏云
王作功
关键词 自然权重设计权重动态最大均值逼近标准化方法实际权重非线性评价方法    
Abstract

[Objective] This paper explores the implicit natural weight issues facing the scientific and technology review indexes, and then proposes a method to address them. [Methods] First, we analyzed data from the JCR2016 mathematics journals with the help of TOPSIS method, aiming to find the influence of natural weights on the nonlinear evaluation method. Then, we proposed a method increasing the dynamic maximum mean to the standardized level, aiming to eliminate the impacts. [Results] We found that the natural weights posed significant effects to the Nonlinear Evaluation methods. For the weighted method, the design weights, the natural weights and the evaluation methods all affected the actual weights. For the non-weighted method, the natural weights and the evaluation methods affected the actual weights. Eliminating the natural weights could effectively reduce the influence of the evaluation method on the actual weights, which helps the design weights play a bigger role. The distribution of index data also affected the actual weights. [Limitations] The proposed method is still an approximation algorithm, which could not yield the exactly equal means. [Conclusions] To achieve the fair review for the science and technology products, we must pay attention to the natural weights issues, which is a systematic error.

Key wordsNatural Weights    Design Weights    Standardization Method of Approximate    Dynamic Maximum Mean    Actual Weights    Nonlinear Evaluation Method
收稿日期: 2017-11-14      出版日期: 2018-07-11
ZTFLH:  G302  
基金资助:*本文系教育部人文社会科学研究规划基金项目“协同创新深度的影响机制与对策研究”(项目编号: 17YJA630125)和浙江省哲学社会科学规划课题“浙江省全面创新的评价体系与推进路径研究”(项目编号: 17NDJC107YB)的研究成果之一
引用本文:   
俞立平, 宋夏云, 王作功. 自然权重对非线性科技评价的影响及纠正研究*——以TOPSIS方法评价为例[J]. 数据分析与知识发现, 2018, 2(6): 48-57.
Yu Liping,Song Xiayun,Wang Zuogong. Impacts and Corrections of Natural Weight on Nonlinear Sci-tech Reviews——Case Study of TOPSIS Method. Data Analysis and Knowledge Discovery, 2018, 2(6): 48-57.
链接本文:  
http://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2017.1124      或      http://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2018/V2/I6/48
评价指标 均值 极大值 极小值 标准差 偏度 峰度 JB检验 概率
总被引频次 7.252 100 0.521 11.467 4.016 23.612 5994.545 0.000
影响因子 16.785 100 4.710 12.098 3.243 17.460 3076.852 0.000
他引影响因子 15.646 100 2.675 12.098 3.308 17.983 3286.246 0.000
影响因子百分位 47.073 100 0.810 27.627 0.146 1.915 15.481 0.000
5年影响因子 21.049 100 5.773 14.642 2.973 14.145 1954.897 0.000
特征因子 9.292 100 0.401 13.318 3.668 19.406 3956.367 0.000
标准特征因子 9.292 100 0.411 13.317 3.668 19.406 3956.613 0.000
论文影响分值 13.795 100 1.164 14.562 3.484 18.406 3502.304 0.000
即年指标 7.526 100 0.000 8.757 5.200 47.209 25267.110 0.000
被引半衰期 21.196 100 0.000 27.304 0.995 2.651 50.026 0.000
引用半衰期 0.680 100 0.000 6.839 12.356 166.623 335445.512 0.000
  JCR2016数学期刊指标描述统计
  权重分类体系
  权重之间的关系
  标准化过程
  本文研究框架
评价指标 设计权重 自然权重 实际权重
他引影响因子X1 0.60 0.486 0.791
特征因子X2 0.25 0.289 0.161
总被引频次X3 0.15 0.225 0.048
  加权TOPSIS三种权重比较
评价指标 设计权重 自然权重 实际权重
他引影响因子X1 0.60 0.333 0.523
特征因子X2 0.25 0.333 0.264
总被引频次X3 0.15 0.333 0.213
  消除自然权重后加权TOPSIS权重比较
期刊名称 消除前
评价值
排名 消除后
评价值
排名
KINET RELAT MOD 0.178 48 0.194 50
RANDOM STRUCT ALGOR 0.172 51 0.194 51
INDIANA U MATH J 0.165 53 0.192 52
J COMB THEORY A 0.161 56 0.191 53
COMMUN CONTEMP MATH 0.170 52 0.188 54
ADV CALC VAR 0.173 49 0.187 55
COMBINATORICA 0.161 55 0.182 56
MEM AM MATH SOC 0.153 59 0.181 57
J PURE APPL ALGEBRA 0.147 66 0.180 58
J ANAL MATH 0.158 58 0.179 59
ERGOD THEOR DYN SYST 0.149 63 0.176 60
CAN J MATH 0.152 60 0.175 61
COMMUN NUMBER THEORY 0.160 57 0.173 62
EUR J COMBIN 0.140 74 0.171 63
SCI CHINA MATH 0.142 72 0.168 64
MOSC MATH J 0.150 61 0.167 65
POTENTIAL ANAL 0.146 67 0.167 66
ELECTRON J COMB 0.134 78 0.166 67
J EVOL EQU 0.149 62 0.166 68
ELECTRON J DIFFER EQ 0.138 75 0.165 69
RUSS MATH SURV+ 0.141 73 0.165 70
ALGEBR NUMBER THEORY 0.142 70 0.165 71
INTERFACE FREE BOUND 0.149 64 0.163 72
PAC J MATH 0.128 87 0.163 73
REND LINCEI-MAT APPL 0.148 65 0.162 74
REV MAT COMPLUT 0.146 68 0.161 75
B LOND MATH SOC 0.129 83 0.159 76
ADV NONLINEAR STUD 0.142 71 0.159 77
J NUMBER THEORY 0.125 91 0.158 78
J GEOM ANAL 0.133 79 0.158 79
  自然权重消除前后加权TOPSIS评价结果比较
评价指标 自然权重 实际权重
他引影响因子X1 0.486 0.657
特征因子X2 0.289 0.226
总被引频次X3 0.225 0.117
  自然权重消除前非加权TOPSIS两种权重比较
评价指标 自然权重 实际权重
他引影响因子X1 0.333 0.320
特征因子X2 0.333 0.322
总被引频次X3 0.333 0.358
  自然权重消除后非加权TOPSIS两种权重比较
期刊名称 消除前
评价值
排名 消除后
评价值
排名
MEM AM MATH SOC 0.142 53 0.189 50
EUR J COMBIN 0.141 54 0.189 51
NUMER LINEAR ALGEBR 0.153 48 0.188 52
P ROY SOC EDINB A 0.146 52 0.185 53
J NUMBER THEORY 0.133 61 0.183 54
ERGOD THEOR DYN SYST 0.135 59 0.182 55
RANDOM STRUCT ALGOR 0.141 55 0.180 56
B LOND MATH SOC 0.125 68 0.174 57
CAN J MATH 0.130 63 0.174 58
ELECTRON J DIFFER EQ 0.125 67 0.172 59
SCI CHINA MATH 0.128 65 0.171 60
RUSS MATH SURV+ 0.125 70 0.170 61
CALCOLO 0.146 51 0.169 62
COMBINATORICA 0.129 64 0.168 63
ADV DIFFERENTIAL EQU 0.138 57 0.168 64
SEL MATH-NEW SER 0.137 58 0.168 65
MATH RES LETT 0.121 71 0.168 66
J COMB THEORY B 0.117 74 0.167 67
J INEQUAL APPL 0.115 76 0.167 68
J INST MATH JUSSIEU 0.138 56 0.165 69
MATH NACHR 0.115 77 0.165 70
COMMUN CONTEMP MATH 0.131 62 0.164 71
J ANAL MATH 0.125 69 0.164 72
KINET RELAT MOD 0.134 60 0.163 73
COMMUN PUR APPL ANAL 0.114 78 0.159 74
CR MATH 0.107 92 0.159 75
J APPROX THEORY 0.112 81 0.159 76
SB MATH+ 0.108 88 0.157 77
DISCRETE COMPUT GEOM 0.107 93 0.157 78
ALGEBR NUMBER THEORY 0.118 72 0.157 79
  自然权重消除前后加权TOPSIS评价结果比较
[1] Vinkler P.Introducing the Current Contribution Index for Characterizing the Recent,Relevant Impact of Journals[J]. Scientometrics, 2008, 79(2): 409-420.
doi: 10.1007/s11192-009-0427-x
[2] Adler R, Ewing J, Taylor P.Citation Statistics: A Report from the International Mathematical Union (IMU) in Cooperation with the International Council of Industrial and Applied Mathematics (ICIAM) and the Institute of Mathematical Statistics (IMS)[J]. Statistical Science, 2009, 24(1): 1-14.
doi: 10.1214/09-STS285
[3] Seglen P O.The Skewness of Science[J]. Journal of the Association for Information Science and Technology, 1992, 43(9): 628-638.
[4] 俞立平, 潘云涛, 武夷山. 科技评价中不同客观评价方法权重的比较研究[J]. 科技管理研究, 2009(7): 148-150.
doi: 10.3969/j.issn.1000-7695.2009.08.047
[4] (Yu Liping, Pan Yuntao, Wu Yishan.Comparative Research on Weight of Different Objective Evaluation Methods in Scientific and Technological Evaluation[J]. Science and Technology Management Research, 2009(7): 148-150.)
doi: 10.3969/j.issn.1000-7695.2009.08.047
[5] 俞立平, 刘爱军. 主成分与因子分析在期刊评价中的改进研究[J]. 情报杂志, 2014, 33(12): 94-98.
doi: 10.3969/j.issn.1002-1965.2014.12.018
[5] (Yu Liping, Liu Aijun.The Misunderstandings and Optimization of Principal Component and Factor Analysis in Journal Evaluation[J]. Journal of Intelligence, 2014, 33(12): 94-98.)
doi: 10.3969/j.issn.1002-1965.2014.12.018
[6] 何强. 群组评价中指标最优权重设计[J]. 统计研究, 2011, 28(8): 99-102.
doi: 10.3969/j.issn.1002-4565.2011.08.015
[6] (He Qiang.Optimal Weighting Design of Indicators in Group Evaluation[J]. Statistical Research, 2011, 28(8): 99-102.)
doi: 10.3969/j.issn.1002-4565.2011.08.015
[7] 苏术锋. 客观评价法中的数据差异赋权有效性及实证[J]. 统计与决策, 2015(21): 74-76.
doi: 10.13546/j.cnki.tjyjc.2015.21.021
[7] (Su Shufeng.Validity and Demonstration of Data Discrepancy Empowerment in Objective Evaluation Method[J]. Statistics and Decision, 2015(21): 74-76.)
doi: 10.13546/j.cnki.tjyjc.2015.21.021
[8] 邹树梁, 武良鹏. 混合多属性决策问题中的权重研究[J]. 运筹与管理, 2017, 26(1): 76-81.
[8] (Zou Shuliang, Wu Liangpeng.Research on the Weight of Hybrid Multiple Attribute Decision-making Problem[J]. Operations Research and Management Science, 2017, 26(1): 76-81.)
[9] 王化中, 强凤娇, 陈晓暾. 模糊综合评价中权重与评价原则的重新确定[J]. 统计与决策, 2015(8): 24-27.
[9] (Wang Huazhong, Qiang Fengjiao, Chen Xiaotun.Redetermination of Weight and Evaluation Principle in Fuzzy Comprehensive Evaluation[J]. Statistics and Decision, 2015(8): 24-27.)
[10] Hagerty M R, Land K C.Constructing Summary Indices of Quality of Life: A Model for the Effect of Heterogeneous Importance Weights[J]. Sociological Methods and Research, 2007, 35(4): 455-496.
doi: 10.1177/0049124106292354
[11] Kahneman D, Tversky A.Prospect Theory: An Analysis of Decision Under Risk[J]. Econometrica, 1979, 47(2): 263-292.
doi: 10.2307/1914185
[12] Edwards W, Barron F H.SMART and SMARTER: Improved Simple Methods for Multiattribute Utility Measurement[J]. Organizational Behavior and Human Decision Processes, 1994, 60(3): 306-325.
doi: 10.1006/obhd.1994.1087
[13] 周志远, 沈固朝. 粗糙集理论在情报分析指标权重确定中的应用[J]. 情报理论与实践, 2012, 35(9): 61-65.
[13] (Zhou Zhiyuan, Shen Guchao.Application of Rough Set Theory in Determining the Weight of Information Analysis Index[J]. Information Studies: Theory & Practice, 2012, 35(9): 61-65.)
[14] 曹秀英, 梁静国. 基于粗集理论的属性权重确定方法[J]. 中国管理科学, 2002, 10(5): 98-100.
doi: 10.3321/j.issn:1003-207X.2002.05.020
[14] (Cao Xiuying, Liang Jingguo.The Method of Ascertaining Attribute Weight Based on Rough Sets Theory[J]. Chinese Journal of Management Science, 2002, 10(5): 98-100.)
doi: 10.3321/j.issn:1003-207X.2002.05.020
[15] 王祖和, 亓霞. 多资源均衡的权重优选法[J]. 管理工程学报, 2002, 16(3): 91-93.
doi: 10.3969/j.issn.1004-6062.2002.03.022
[15] (Wang Zuhe, Qi Xia.The Weight Optimal Choice Method of Multi-Resource Leveling[J]. Journal of Industrial Engineering/Engineering Management, 2002, 16(3): 91-93.)
doi: 10.3969/j.issn.1004-6062.2002.03.022
[16] 周辉, 鲁燕飞, 王黔英, 等. 基于信息粒度的属性权重确定方法[J]. 统计与决策, 2006(20): 134-136.
[16] (Zhou Hui, Lu Yanfei, Wang Qianying, et al.Attribute Weight Determination Method Based on Information Granularity[J]. Statistics and Decision, 2006(20): 134-136.)
[17] 何立华, 王栎绮, 张连营. 基于聚类的多属性群决策专家权重确定方法[J]. 运筹与管理, 2014, 23(6): 65-72.
doi: 10.3969/j.issn.1007-3221.2014.06.011
[17] (He Lihua, Wang Liqi, Zhang Lianying.A Method for Determining the Experts’ Weights of Multi-Attribute Group Decision-Making Based on Clustering Analysis[J]. Operations Research and Management Science, 2014, 23(6): 65-72.)
doi: 10.3969/j.issn.1007-3221.2014.06.011
[18] 张立军, 邹琦. 基于路径系数权重的科技成果奖励评价模型[J]. 科技管理研究, 2008(5): 102-103.
[18] (Zhang Lijun, Zou Qi.Evaluation Model of Scientific and Technological Achievements Based on the Weight of Path Coefficient[J]. Science and Technology Management Research, 2008(5): 102-103.)
[19] 陈亮, 成榕, 岳立柱. 众里取大规则下由频率确定属性权重的方法[J]. 统计与决策, 2017(8): 63-66.
[19] (Chen Liang, Cheng Rong, Yue Lizhu.Method of Frequency Determining the Attribute Weights Based on the Take-the-Biggest- of-All Rule[J]. Statistics and Decision, 2017(8): 63-66.)
[20] 岳立柱, 闫艳. 基于序数信息的属性权重确定方法[J]. 统计与决策, 2015(13): 78-80.
doi: 10.13546/j.cnki.tjyjc.2015.13.020
[20] (Yue Lizhu, Yan Yan.Research on Attribute Weight Determination Method Based on Ordinal Information[J]. Statistics and Decision, 2015(13): 78-80.)
doi: 10.13546/j.cnki.tjyjc.2015.13.020
[21] 傅蓉. 平衡计分卡指标权重前后不一致现象研究[J]. 金融论坛, 2011(9): 71-74.
[21] (Fu Rong.A Study of Inconsistency of Initial Weight with Result Weight in Balanced Scorecard[J]. Finance Forum, 2011(9): 71-74.)
[22] 俞立平, 刘爱军. 期刊评价中TOPSIS的漏洞研究——权重单调性[J]. 情报杂志, 2014, 33(11): 131-135, 192.
doi: 10.3969/j.issn.1002-1965.2014.11.024
[22] (Yu Liping, Liu Aijun.Study on TOPSIS’s Vulnerability in Journal Evaluation - Weights Monotony[J]. Journal of Intelligence, 2014, 33(11): 131-135, 192.)
doi: 10.3969/j.issn.1002-1965.2014.11.024
[1] 俞立平, 潘云涛, 武夷山. 学术期刊非线性评价方法的检验与修正研究[J]. 现代图书情报技术, 2011, 27(7/8): 110-115.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn