Abstract：Key core technology identification is of great significance for promoting the development of science and technology in China. This paper proposes an LDA-Prophet based process for the identification and foresight of key core technologies. Firstly,the paper analyzes the patent materials in related fields through LDA to identify the scope of core technologies. Then key core technologies are identified through a four-dimensional identification framework. Finally,key core technology prediction is carried out through Prophet time series model. Taking the field of chip materials as an example,the current topics of key core technologies are identified and topics for the next three years are predicted. This verifies that this process can effectively identify and predict key core technologies,and can provide useful references for related fields.
蔡鸿宇, 马亚星, 李明, 石进. 基于LDA-Prophet模型的关键核心技术识别与预测研究——以芯片材料领域为例[J]. 中国科技论坛, 2026(5): 74-83.
Cai Hongyu, Ma Yaxing, Li Ming, Shi Jin. Research on Identification and Prediction of Key Core Technologies Based on LDA-Prophet Modeling——Taking the Field of Chip Materials as an Example. , 2026(5): 74-83.
[1] 吕鲲,罗星雨,靖继鹏.融合CFDP和CPM分析的关键核心技术识别及其路径分析:以芯片光刻领域为例[J].图书情报工作,2024,68(16):75-89. [2] 张玉臣,谭礼.关键核心技术的概念界定、特征辨析及突破路径[J].中国科技论坛,2023(2):20-29. [3] 李露,魏巍,文庭孝.多维视角下的关键核心技术识别:概念辨析与研究进展[J].世界科技研究与发展,2024,46(3):371-385. [4] 方曦,崔梁雨,刘云.基于专利特征的人工智能领域核心技术识别及发展趋势分析:层次分析(AHP)-熵权法[J].科技管理研究,2024,44(2):35-44. [5] 孙彩虹,蔡鸿宇,石进,等.专利应用视角下的关键核心技术识别研究[J].情报理论与实践,2025,48(1):121-130. [6] 冉从敬,田文芳.融合SVM-LDA与加权相似度的潜在新兴技术识别研究:以人工智能领域为例[J].情报学报,2024,43(5):563-574. [7] 田雪姣,鲍新中,杨大飞,等.基于熵权-TOPSIS-德尔菲法的核心技术识别研究:以芯片产业技术为例[J].情报杂志,2022,41(8):69-74,86. [8] LEE C.Logistics technology forecasting framework using patent analysis for technology roadmap[J].Sustainability,2022,14(9):5430. [9] SMITH M,AGRAWAL R.A comparison of time series model forecasting methods on patent groups[J].MAICS 2015,1353:167-173. [10] LIU Z,XIANG X,FENG J.Tracing evolutionary trajectory of charging technologies in electric vehicles:Patent citation network analysis[J].Environment,Development and Sustainability,2024,26(5):12789-12813. [11] 周小琴,蔡鸿宇,石进,等.融合专利SAO特征的早期专利价值评估方法研究[J].情报杂志,2024,43(10):176-186. [12] BLEI D M,NG A Y,JORDAN M I.Latent dirichlet allocation[J].Journal of Machine Learning Research,2003,3(1):993-1022. [13] 蔡鸿宇,王向锋,徐宗煌,等.高价值专利有何特征:基于模糊集定性比较分析的组态效应研究[J].情报科学,2024,42(4):167-174,183. [14] 白敬毅,颜端武,陈琼.基于主题模型和曲线拟合的新兴主题趋势预测研究[J].情报理论与实践,2020,43(7):130-136,193. [15] 李瑞茜,陈向东.基于专利共类的关键技术识别及技术发展模式研究[J].情报学报,2018,37(5):495-502. [16] 张文君,顾行发,陈良富,等.基于均值-标准差的K均值初始聚类中心选取算法[J].遥感学报,2006(5):715-721. [17] 李慧,玄洪升.专利视角下融合多属性的技术创新主题挖掘方法:以芯片领域专利为例[J].图书情报工作,2020,64(11):96-107. [18] TAYLOR S J,LETHAM B.Forecasting at scale[J].The American Statistician,2018,72(1):37-45. [19] 杨大飞,杨武,田雪姣,等.基于专利数据的核心技术识别模型构建及实证研究[J].情报杂志,2021,40(2):47-54.
2026, 
