基于多源信息融合的产业链关键核心技术主题识别研究——以人工智能领域为例  

Research on Topic Recognition of Key Core Technology in Industrial Chain Based on Multi-source Information Fusion:Taking AI as an Example

作　　者：李维思[1] 谭力铭 章国亮 文晓芬[1] 廖婷[1] 史敏[2] Li Weisi;Tan Liming;Zhang Guoliang;Wen Xiaofen;Liao Ting;Shi Min(Hunan Institute of Scientific and Technological Information,Changsha,410001;Hunan Agricultural University,Changsha,410128)

机构地区：[1]湖南省科学技术信息研究所,长沙410001 [2]湖南农业大学,长沙410128

出　　处：《信息资源管理学报》2022年第1期116-126,共11页Journal of Information Resources Management

基　　金：国家自然科学基金项目“基于专利语义分析的潜在竞争对手识别方法研究”(71774052)研究成果之一;湖南省自然科学基金项目“基于多源信息融合处理的湖南省人工智能领域技术预测研究”(2020JJ4389)成果之一。

摘　　要：新一轮科技革命蓬勃兴起,全球科技竞争态势日趋激烈,在科技管理过程中,科学研判产业创新趋势、甄选布局关键技术攻关项目意义十分重大。本研究从产业链关键核心技术概念特征出发,融合处理科技论文、项目、成果、专利等多源信息,研究开发了基于LDA主题模型文本挖掘的关键核心技术识别方法,并以我国人工智能产业为例,在产业链各环节分析基础上,开展关键核心技术识别与技术预测,为政府和企业提供具有前瞻性、时效性和专业化的情报分析和决策支撑。A new round of scientific and technological revolution is booming,and global competition in science and technology is increasingly encouraging.In the process of science and technology management,it is very important to scientifically judge the trend of industrial innovation and select key technology projects for layout.This paper elaborates the concept characteristics of the key core technology of the industrial chain process multi-source information such as scientific papers,projects,achievements and patents and establish the key technology recognition method based on LDA model.Then the multi-source scientific and technological information in AI is collected.The key technologies are forecasted based on the AI industrial chain analysis.It is timely,forward-looking and professional for government and enterprises strategic decision.

关 键 词：产业链 关键核心技术 多源信息融合 技术识别 人工智能 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]