基金项目计划学部交叉及对前沿分布的影响研究——以美国NSF数据中AI领域为例  

Frontier Influence of the Inter-Directorate in Project Planning:Case Studies of AI Field in NSF Data

作　　者：范丽鹏 王曰芬[1,2,3] 岑咏华 杨洁 Fan Lipeng;Wang Yuefen;Cen Yonghua;Yang Jie(School of Economics&Management,Nanjing University of Science&Technology,Nanjing 210094;Management School of Tianjin Normal University,Tianjin 300387;Institute for Big Data Science,Tianjin Normal University,Tianjin 300387)

机构地区：[1]南京理工大学经济管理学院,南京210094 [2]天津师范大学管理学院,天津300387 [3]天津师范大学大数据科学研究院,天津300387

出　　处：《情报学报》2022年第9期956-966,共11页Journal of the China Society for Scientific and Technical Information

基　　金：国家自然科学基金应急管理项目“人工智能领域研究前沿探测与决策支持”(61842602)。

摘　　要：本文旨在融合学者参与度的基础上,测算和分析基金项目计划的学部交叉度,并识别前沿型项目计划,从项目资助专项层面上探究学部交叉对项目计划资助与前沿型项目计划分布的影响。本文首先以项目计划为研究对象,为融合项目计划的学部多样性和不同学者对同一项目计划各学部参与的均衡性,引入Rao-Stirling指标,将项目计划分为学部高交叉、中交叉、低交叉和不交叉类;其次,依据项目计划的资助强度和资助趋势,结合学部交叉特性,将项目计划分为不同交叉水平的前沿型、热点型、潜在型和衰退型;最后,对高交叉前沿型项目计划进行具体分析。研究结果表明,以美国NSF (National Science Foundation)资助的AI (artificial intelligence)领域为例,NSF资助的项目计划在学部交叉和不交叉类的数目上分布较为均衡,但对交叉类项目计划的平均资助强度远高于不交叉类型;在前沿分布上,不交叉类的项目计划多为潜在型,而学部高交叉类和低交叉类的前沿型占比相对较高,且对前沿型项目计划的资助趋势向高交叉类倾斜;高交叉类的前沿型项目计划多倾向于神经与认知系统、自然与人类系统、信息智能系统等方面的研究。The purpose of this study is to calculate and analyze the inter-directorate degree in project planning;it further identifies the front of project planning for exploring the inter-directorate influence and the frontier distribution of project funding while considering the balance of researchers taking part in programs with different directorates and how many directorates the program involves.First,we use the Rao-Stirling diversity indicator to calculate the inter-directorate degree of the programs before dividing them into high-crossing,medium-crossing,low-crossing,and non-crossing types.According to the funding trend and intensity of the program,we then construct the indicators of frontier identification and divide the programs into cutting-edge,hot-spot,potential,and decline types with different inter-directorate degrees.Finally,we analyze the inter-directorate influence on the distribution of frontier programs and elaborate more on the cutting-edge programs in high-crossing types.The research results show that,for National Science Foundation(NSF) data in the artificial intelligence(AI) field,the number of crossing programs was approximately equal to that of non-crossing programs while the average funding of cross-type programs was much higher than that of non-crossing programs.From the perspective of the distribution in frontier programs,the non-crossing programs were generally the potential type,while high-crossing and low-crossing types were more likely to be cutting-edge programs.The funding trend was higher for high-crossing type than that for other crossing types.Moreover,it was observed that the fronts of high-crossing programs tend to focus on neural and cognitive systems,natural and human systems,and information intelligence systems.

关 键 词：学部交叉 前沿识别 Rao-Stirling指数 美国国家科学基金 人工智能 

分 类 号：G203[文化科学—传播学]