摘要
评估科学家学术影响力的传统方法只考虑科学家的产出(论文数量、引用总数等)而不考虑科学家的投入。提出一种基于超效率DEA的科学家学术影响力模型,将合作科学家数量和合作机构数量作为产出要素,论文总数和引用总数作为投入要素。在APS数据集上的实验结果表明,诺贝尔奖科学家的学术影响力比非诺贝尔奖科学家的学术影响力高,超效率DEA模型计算出的AUC值为0.795 6,比传统指标中最高总引用量指标提高了8.75%,对评价科学家的学术影响力有十分重要的借鉴意义。
The current methods of quantifying and assessing academic Influence Analysis of scientists only take into account the outputs of scientists,regardless the fact that the input factors are different for each scientist.In this paper,taking into account the input and output factors measured by the number of cooperating scientists,the number of cooperating institutions,total number of papers and total number of citations,we present a model to evaluate the performance of scientists.The method on the APS dataset can more accurately identified Nobel Prize-winning scientists than the published articles,total citations,I10 index,and H index.The AUC value of this model was 0.7956,which was 8.75% higher than total number of citations.The work of this paper is of great significance for quantifying the academic influence of scientists.
作者
陈清文
郭强
刘建国
CHEN Qing-wen;GUO Qiang;LIU Jian-guo(Complex Systems Science Research Center,University of Shanghai for Science and Technology,Shanghai 200093,China;Institute of Financial Technology Laboratory,Shanghai University of Finance and Economics,Shanghai 200433,China)
出处
《软件导刊》
2019年第5期155-157,161,共4页
Software Guide
基金
国家自然科学基金项目(61773248)
