摘要
本文将权重分为设定权重与实际权重,指出线性评价方法设定权重与设计权重是相等的,而非线性评价方法两者并不一致,实际权重可以用评价结果与评价指标进行岭回归,然后将回归系数标准化后得到。基于JCR经济学期刊数据,通过对加权TOPSIS的研究表明,加权TOPSIS评价方法不具有权重单调性,加权TOPSIS评价压低设定权重; TOPSIS评价方法具有较强的权重鲁棒性;在此基础上,提出了分子加权TOPSIS方法,解决了权重单调性问题。
In this paper,the weight was divided into the set weight and the actual weight,there was no difference between the weights set up or designed under the linear evaluation method,or they were not consistent,The Ridge Regression could be carried out with the evaluation results and evaluation indexes,and then the regression coefficients were standardized,the result was actual weight. The weighted TOPSIS,based on the data of JCR Journal of economics,showed that the evaluation method of weighted TOPSIS did not possess weight monotonicity,set weight could be depressed by the evaluation method of weighted TOPSIS; Evaluation method of weighted TOPSIS had strong robustness for weight. As a result,this paper put forward the method of molecular weighted TOPSIS,which solved the problem for weight monotonicity.
引文
[1] Hwang C,Yoon K. Multiple Attribute Decision Making:Methods and Applications[M]. Berlin:Springer-Verlag,1981:1-50
[2] Shyur H J. COTS Evaluation Using Modified TOPSIS and ANP[J]. Applied Mathematics and Computation,2006,177(1):251-259
[3] Deng H,Yeh C H,Willis R J. Inter-company Comparison Using Modified TOPSIS with Objective Weights[J]. Computers&Operations Research,2000,27(10):963-973.
[4] Yue Z L. An extended TOPSIS for Determining Weights of Decision Makers with Interval Numbers[J]. Knowledge-Based Systems,2011,24(1):146-153.
[5] Xu J,Li Z,Shen W,et al. Benjamin Lev. Multi-attribute Comprehensive Evaluation of Individual Research Output Based on Published Research Papers[J]. Knowledge-Based Systems,2013,(43):135-142.
[6]胡永健,周琼琼,张杰军.基于多属性决策的国家科技基础条件平台运行服务绩效评估研究[J].中国科技论坛,2009,(12):3-7.
[7] Celika M,Kandakoglu A,Er I D. Structuring Fuzzy Integrated Multi-stages Evaluation Model on Academic Personnel Recruitment in MET Institutions[J]. Expert Systems with Applications,2009,(36):6918-6927.
[8] Li Z,Liechty M,Xu J,et al. A Fuzzy Multi-criteria Group Decision Making Method for Individual Research Output Evaluation with Maximum Consensus[J]. Knowledge-Based Systems,2014,(56):253-263.
[9]赵黎明,刘猛.基于熵权TOPSIS的区域科技创新能力评价模型及实证研究[J].天津大学学报:社会科学版,2014,(5):385-390.
[10]皮进修,彭建文,赵清俊.大数据研究领域中科研机构影响力测度研究[J].情报杂志,2016,(7):179-184,125.
[11]王天歌,王金苗,袁红梅.基于专利维度的我国生物医药核心技术的识别与分析[J].情报杂志,2016,(4):112-117.
[12]张夏恒,冀芳.大气科学类中文核心期刊微信公众号满意度评价[J].中国科技期刊研究,2017,(1):47-52.
[13]邵景波,李柏洲,周晓莉.基于加权主成分TOPSIS价值函数模型的中俄科技潜力比较[J].中国软科学,2008,(9):132-139.
[14]石宝峰,迟国泰,章穗.基于矩阵距离时序赋权的科学技术评价模型及应用[J].运筹与管理,2014,(1):166-178.
[15]王映.加权TOPSIS与RSR法在学术期刊影响力综合评价中的应用研究[J].图书情报工作,2013,(2):92-96.
[16]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,(11):131-135,192.
[17] Jahanshahloo G R,Lotfi F H,Izadikhah M. Extension of the TOPSIS Method for Decision-making Problems with Fuzzy Data[J]. Applied Mathematics and Computation,2006,181(2):1544-1551.
[18]陆伟锋,唐厚兴.关于多属性决策TOPSIS方法的一种综合改进[J].统计与决策,2012,(19):38-40.
[19]华小义,谭景信.基于“垂面”距离的TOPSIS法——正交投影法[J].系统工程理论与实践,2014,(1):114-119.
[20]谭春桥.基于区间值直觉模糊集的TOPSIS多属性决策[J].模糊系统与数学,2010,(2):92-97.
[21]孙世岩,邱志明,张雄飞.多属性决策鲁棒性评价的仿真方法研究[J].武汉理工大学学报:信息与管理工程版,2006,(12):58-61.
[22]徐泽水.一种基于目标贴近度的多目标决策方法[J].系统工程理论与实践,2001,(9):101-104.
[23]付巧峰.一种修改的TOPSIS法[J].西北大学学报:自然科学版,2007,(4):531-533.
[24]俞立平,潘云涛,武夷山.学术期刊多属性评价方法的可比性研究[J].编辑学报,2010,(5):381-384.