基于系统辨识的全球科技领军人才流动网络化模型研究

针对全球科技领军人才流动问题,将文献计量学、复杂网络和系统辨识三种分析工具相结合,建立全球科技领军人才跨国流动网络化模型.首先,应用文献计量学方法,给出全球范围内重点科技领域的领军人才的准确定义.提出基于履历分析法收集全球主要国家科技领军人才流动数据的方法:之后,结合代数图论、矩阵理论和复杂网络理论,建立全球科技人才流动网络框架模型,确定框架模型中能够直接推导出的参数和待辨识的未知参数,应用定量化模型描述全球科技人才流动的整体态势;再次,将各个国家的全球科技人才总量按时间段进行收集汇总,建立相应的数据库,基于数据库应用最小二乘方法辨识框架模型中的未知参数,从理论上完成全球科技人才流动网络模型的构建.最后.应用部分领域全球主要国家科技领军人才数据,对模型的有效性进行验证,分析领军人才流动的特点.

Network of global top scientific talent migration based on system identification

This paper investigates international migration for top talent with three kinds of analysis tools—bibliometrics, complex network theory and system identification. First, according to bibliometerics, the clear definition of top talents is given, and the method of collection of historical data about the amounts of top talents of each countries is introduced. Moreover, the framework of international migration is studied from algebraic graph theory, matrix theory and complex network theory. And the relationship between complex network and international migration is discussed. Also, the known and unknown parameters of the proposed framework are distinguished. Based on the model, the overall situation of international migration of top talents can be represented quantitatively. In addition, we collect the residing information of top talents, and establish the corresponding database. Furthermore, by employing the least squares method, this paper proposes two approaches to identify the unknown parameters, and the obtained identification results are testified based on statistics theory. Finally, the migration of top talents between some major and typical countries is discussed according to proposed method, and the analytic results and the effectiveness are testified.