| 基于专利文献的多指标新兴技术识别研究 |
| |
| 引用本文： | 宋欣娜,郭颖,席笑文.基于专利文献的多指标新兴技术识别研究[J].情报杂志,2020,39(6):76-81,88. |
| |
| 作者姓名： | 宋欣娜 郭颖 席笑文 |
| |
| 作者单位： | 北京理工大学 管理与经济学院 北京 100081;中国政法大学 商学院 北京 100088 |
| |
| 基金项目： | 国家自然科学基金;中国政法大学钱瑞升杰出学者支持计划资助项目 |
| |
| 摘 要： | 目的/意义]对新兴技术进行识别,有助于跟踪新兴技术的最新发展动态,为国家、企业、学术界提供最新的情报。方法/过程]以专利数据为基础,利用文献计量学和文本挖掘方法对纳米载药系统领域进行了新兴技术识别。首先,确定了识别体系中的四个指标:新颖性、持久性、社区性、增长性,并为四个指标设定了相应的筛选条件。与前人研究相比,考虑了主题扩散和不同数据集增长模式的差异,使筛选条件更为完善;通过引入新兴分数和LDA主题模型分别得到了新兴术语和新兴主题,实现了在术语和主题两个层面对新兴技术的识别;使用指标验证法对得到的新兴术语进行了验证,并与前人研究方法进行了对比。结果/结论]结果表明,该方法优化了现有的新兴技术识别模型。
|
| 关 键 词： | 新兴技术 专利数据 文献计量学 文本挖掘 LDA主题模型 指标验证法 |
Research on Multi-Indicator Emerging Technology Identification Based on Patent Literature |
| |
| Song Xinna,Guo Ying,Xi Xiaowen.Research on Multi-Indicator Emerging Technology Identification Based on Patent Literature[J].Journal of Information,2020,39(6):76-81,88. |
| |
| Authors: | Song Xinna Guo Ying Xi Xiaowen |
| |
| Institution: | (School of Management & Economics,Beijing Institute of Technology,Beijing 100081;Business School,China University of Political Science and Law,Beijing 100088) |
| |
| Abstract: | Purpose/Significance]The identification of emerging technologies can help to track the latest development dynamics of emerging technologies and provide the latest intelligence information for countries,enterprises and academia.Method/Process]Based on the patent data,this study uses bibliometrics and text mining methods to identify the emerging technologies in the field of Nano-Enabled Drug Delivery(NEDD).First,this study identifies four indicators of the emerging technology identification system:novelty,persistence,community,and growth,and the corresponding screening conditions are set for the four indicators.Compared with previous studies,this study considers the topic diffusion and the difference of different data set growth patterns,so as to improve the screening conditions.Then,by introducing the emergence score and LDA topic model,the emerging terms and emerging topics are obtained respectively,realizing the identification of emerging technologies at the level of terms and topics.Finally,this study uses the index validation method to verify the emerging terms obtained and compares with previous research methods.Result/Conclusion]The results show that this study optimizes the existing emerging technology identification model. |
| |
| Keywords: | emerging technology patent data bibliometrics text mining LDA model index validation method |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
