| 曹京晶,王莹,王烨,陈皓,张雪,张晓夏,朱晓晨.美国科学基金资助热点布局及对我国海洋领域规划的借鉴[J].情报工程,2021,(5):062-074 |
| 美国科学基金资助热点布局及对我国海洋领域规划的借鉴 |
| Distribution of NSF-funded Projects and its Referential Experience to Marine Research Planning of China |
| |
| DOI：10.3772/j.issn.2095-915x.2021.05.006 |
| 中文关键词: 文本挖掘；NSF；海洋科学；前沿识别 |
| 英文关键词: Text Mining; NSF; marine science; frontier identification |
| 基金项目:科技创新战略研究专项（ZLY202111）。 |
| 作者 | 单位 | | 曹京晶 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 王莹 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 王烨 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 陈皓 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 张雪 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 张晓夏 | 科学技术部科技人才交流开发服务中心 北京 100045 | | 朱晓晨 | 科学技术部科技人才交流开发服务中心 北京 100045 |
|
| 摘要点击次数: 1306 |
| 全文下载次数: 970 |
| 中文摘要: |
| [ 目的 / 意义 ] 从世界强国的发展经历看，要想崛起于世界，必先强盛于海洋。我国海洋科技起步较晚，海洋科技创新能力基本处于全球海洋科技竞争格局中的第二梯队，与第一梯队还存在一定的差距。本文旨在通过分析美国国家科学基金会（NSF）对海洋领域的资助情况，对海洋领域的热点方向及布局开展分析研究，以期对我国海洋领域的规划布局提供参考。[ 方法 / 过程 ] 本文主要通过对NSF 近四十年来项目资助数量、资助金额、资助时间等数据维度进行简要说明，并通过 NSF 在海洋领域的样本数据，尝试将 LDA 模型与 BERT 模型相结合，首先采用基于 Bert 的模型将文本向量化，然后基于 LDA（Latent Dirichlet Allocation）算法对文本聚类，以解决提高聚类准确度的文本自动分类任务。尝试分析出海洋领域的重点关键技术及研究热点方向，为我国未来对海洋领域的规划及布局提供支撑。[ 结果 / 结论 ] 通过 NSF 在海洋领域的样本数据，尝试分析出海洋领域的重点关键技术及研究热点方向，为我国未来对海洋领域的规划及布局提供参考。 |
| 英文摘要: |
| [Objective/ Significance] Judging from the development experience of world powers, if they want to rise in the world, they must first become stronger in the ocean. my country’s marine science and technology started relatively late, and the innovation capability of marine science and technology is basically in the second echelon of the global marine science and technology competition, and there is still a certain gap between it and the first echelon. The purpose of this article is to analyze the US NSF’s funding for the marine field, and conduct analysis and research on the hotspot direction and layout of the marine field, in order to provide a reference for the planning and layout of my country’s marine field. [Methods/Process] This paper mainly introduces the data dimensions of NSF’s project funding amount, funding amount, funding time and other data in the past 40 years, and tries to combine the LDA model with the BERT model through the sample data of the NSF in the marine field. , Firstly use the Bert-based model to vectorize the text, and then cluster the text based on the LDA (Latent Dirichlet Allocation) algorithm to solve the task of automatic text classification to improve the accuracy of clustering. Try to analyze the key and key technologies and research hotspots in the marine field to provide support for my country’s future planning and layout in the marine field. [Results/Conclusions] Through the sample data of NSF in the marine field, try to analyze the key key technologies and research hotspots in the marine field, so as to provide reference for my country’s future planning and layout of the marine field. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
