图书情报工作

图书情报工作 >

2025 , Vol. 69 >Issue 23: 69 - 83

DOI: https://doi.org/10.13266/j.issn.0252-3116.2025.23.006

研究论文

中美两国人工智能创新趋势预测研究——基于专利申请数据与分解集成方法

  • 梅国平 ,
  • 欧阳嘉文 ,
  • 张阳 ,
  • 石圆圆
展开
  • 江西师范大学管理科学与工程研究中心 南昌 330022
梅国平，教授，博士，博士生导师；欧阳嘉文，博士研究生，通信作者，E-mail：202043700072@jxnu.edu.cn；张阳，副教授，博士；石圆圆，博士研究生。

收稿日期: 2025-03-28

  修回日期: 2025-06-19

  网络出版日期: 2025-11-20

基金资助

本文系国家社会科学基金重大招标项目“新一代人工智能对中国经济高质量发展的影响、趋向及应对战略研究”（项目编号：20&ZD068）和江西省教育厅研究生创新基金项目“人工智能赋能家居产业高质量发展的机制与路径研究”（项目编号：YC2024-B090）研究成果之一。

收起

A Study on Prediction of Artificial Intelligence Innovation Trend in China and the United States: Based on Patent Application Data and Decomposition Integration Method

  • Mei Guoping ,
  • Ouyang Jiawen ,
  • Zhang Yang ,
  • Shi Yuanyuan
Expand
  • Management Science and Engineering Research Center, Jiangxi Normal University, Nanchang 330022
Mei Guoping, Professor, PhD, Doctoral Supervisor; Ouyang Jiawen, Doctoral Candidate, Corresponding Author, E-mail: 1121369179@qq.com; Zhang Yang, Associate Professor, PhD; Shi Yuanyuan, Doctoral Candidate.

Received date: 2025-03-28

  Revised date: 2025-06-19

  Online published: 2025-11-20

Supported by

This work is supported by the major program of National Social Science Fund of China, titled “Research on the Influence, Trend and Response Strategy of the New Generation of AI on the High-Quality Development of China’s Economy” (20&ZD068); and the Jiangxi Provincial Department of Education Graduate Student Innovation Fund Project of China, titled “Research on the Mechanism and Path of Artificial Intelligence Enabling the High-Quality Development of the Home Furnishing Industry” (YC2024-B090).

Fold

摘要

[目的/意义] 构建“分解—集成”预测框架，分析中美两国人工智能创新趋势及其关键特征，对把握全球人工智能技术格局演变具有重要的参考价值。[方法/过程] 基于人工智能发明专利申请数据，系统比较8种分解方法与8种预测模型的组合，识别出VMD-LSTM模型的显著优势，进而应用此模型预测和比较两国人工智能创新的未来趋势，并对VMD分解结果进行特征解读。[结果/结论] 预测结果揭示出未来中国人工智能专利申请量将展现超越美国的增长潜力，这与2025年“DeepSeek”技术的迅速破圈引发人工智能创新热潮相吻合。分解结果显示：两国在2012年具有相似的增长加速趋势；中国自2015年专利申请量超越美国后，其创新活动的波动特征与国家政策的阶段性演进呈现共变关系；面对疫情冲击，中国人工智能创新展现出相较美国更强劲的恢复动能。基于研究结果，对中国人工智能发展提出夯实前瞻布局、优化政策组合、强化趋势监测等建议。

关键词： 人工智能; 中美对比; 趋势预测; 分解集成; 机器学习

本文引用格式

梅国平 , 欧阳嘉文 , 张阳 , 石圆圆 . 中美两国人工智能创新趋势预测研究——基于专利申请数据与分解集成方法[J]. 图书情报工作, 2025 , 69(23) : 69 -83 . DOI: 10.13266/j.issn.0252-3116.2025.23.006

Abstract

[Purpose/Significance] This paper constructs a “decomposition-integration” prediction framework to analyze the innovation trends and key characteristics of artificial intelligence in China and the United States, which is of great reference value for grasping the evolution of global AI technology. [Method/Process] Based on the data of AI invention patent applications, the system compared the combination of eight decomposition methods and eight prediction models to identify the significant advantages of VMD-LSTM model, and then applied this model to predict and compare the future of AI innovation in the two countries, and interpreted the characteristics of VMD decomposition results. [Result/Conclusion] The prediction results indicate that China’s future AI patent applications will show growth potential surpassing that of the United States. The decomposition results reveal that the long-term growth trends of both countries align with breakthroughs in deep learning. Since 2015, when the patent application volume in China significantly surpassed that of the United States, the fluctuation patterns of China’s innovation activities have shown significant co-movement with the phased evolution of national policies. In the face of the pandemic, China’s AI innovation has demonstrated stronger recovery momentum. Based on these findings, recommendations are made for China’s AI development, including strengthening forward-looking planning, optimizing policy combinations, and enhancing trend monitoring.

Key words： artificial intelligence; Sino-American comparison; trend prediction; decomposition integration; machine learning

参考文献

[1] WIPO. WIPO technology trends 2019: Artificial intelligence[R]. Geneva: World Intellectual Property Organization, 2019.
[2] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis[J]. Proceedings of the royal society of London. Series A: mathematical, physical and engineering sciences, 1998, 454(1971): 903-995.
[3] 陈楠, 蔡跃洲. 人工智能、承接能力与中国经济增长——新“ 索洛悖论” 和基于AI专利的实证分析[J]. 经济学动态, 2022(11): 39-57. (CHEN N, CAI Y Z. Al, accommodating capacity and economic growth in China: the new “Solow Paradox” and an empirical analysis using the Al patent data [J]. Economic perspectives, 2022(11): 39-57.)
[4] 尹志锋, 曹爱家, 郭家宝, 等. 基于专利数据的人工智能就业效应研究——来自中关村企业的微观证据[J]. 中国工业经济, 2023(5): 137-154. (YIN Z F, CAO A J, GUO J B, et al. A study of the impact of artificial intelligence on employment based on patents data: evidence from Zhongguancun enterprises [J]. China industrial economics, 2023(5): 137-154.)
[5] 姚加权, 张锟澎, 郭李鹏, 等. 人工智能如何提升企业生产效率？ ——基于劳动力技能结构调整的视角[J]. 管理世界, 2024, 40(2): 101-116, 133, 117-122. (YAO J Q, ZHANG K P, GUO L P, et al. How does artificial intelligence improve firm productivity? Based on the perspective of labor skill structure adjustment[J]. Journal of management world, 2024, 40(2): 101- 116, 133, 117-122.)
[6] 陈梁, 罗敏, 袁润. 新兴技术发展演化的时间特性分析与应用研究——以医疗机器人为例[J]. 情报杂志, 2023, 42(4): 117- 124. (CHEN L, LUO M, YUAN R. Temporal characterization and application study of the development and evolution of emerging technologies: an example of medical robotics[J]. Journal of intelligence, 2023, 42(4): 117-124.)
[7] 陈鑫鑫, 陈中天, 罗福凯. 政府投资基金对企业创新的动态效应研究[J]. 会计与经济研究, 2022, 36(6): 20-37. (CHEN X X, CHEN Z T, LUO F K. Research on the dynamic effect of government investment funds on enterprise innovation[J]. Accounting and economics research, 2022, 36(6): 20-37.)
[8] 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. (RAN C J, TIAN W F. Identification of potential emerging technologies by fusing SVM-LDA and weighted similarity: taking the field of artificial intelligence as an example [J]. Journal of the China Society for Scientific and Technical Information, 2024, 43(5): 563-574.)
[9] 杨锡怡, 贾佳, 周小宇, 等. 中美两国人工智能头部企业研发和创新的比较分析与启示[J]. 中国科学院院刊, 2024, 39(6): 1084-1096. (YANG X Y, JIA J, ZHOU X Y, et al. Comparative analysis and insights into R&D mode of top artificial intelligence companies in China and the US[J]. Bulletin of Chinese Academy of Sciences, 2024, 39(6): 1084-1096.)
[10] 王丹, 周潇, 赵捧未, 等. 基于离群点视角的颠覆性专利预测研究[J]. 图书情报工作, 2024, 68(5): 74-86. (WANG D, ZHOU X, ZHAO P W, et al. Research on disruptive patent prediction based on outlier perspective[J]. Library and information service, 2024, 68(5): 74-86.)
[11] 苗红, 李男, 吴菲菲, 等. 基于机器学习的医学影像人工智 能领 域技 术融 合预 测[J]. 情报 杂志, 2022, 41(6): 126- 134. (MIAO H, LI N, WU F F, et al. Machine-learning-based technology convergence forecasting for medical imaging artificial intelligence [J]. Journal of intelligence, 2022, 41(6): 126-134.)
[12] 谢晓燕, 韦学婷, 王霖. 基于指数平滑法的呼、包、鄂三角区物流需求量预测[J]. 干旱区资源与环境, 2013, 27(1): 58-62. (XIE X Y, WEI X T, WANG L. Prediction of logistics demand based on exponential smoothing model for Hohhot, Baotou and Ordos [J]. Journal of arid land resources and environment, 2013, 27(1): 58-62.)
[13] 李洪晨, 张志强. 全国国民图书阅读率会超过60% 吗？ ——基于ARIMA模型的全国国民阅读调查预测研究[J]. 图书情报工作, 2023, 67(9): 72-80. (LI H C, ZHANG Z Q. Will the national book reading rate exceed 60%?A study of national reading survey prediction based on ARIMA model [J]. Library and information service, 2023, 67(9): 72-80.)
[14] 张婷. CPI的SARIMA模型与X-12季节调整模型对比预测分析[J]. 经济问题, 2014(12): 37-41. (ZHANG T. Comparison of predictive analysis of CPI based on SARIMA and X-12 model [J]. On economic problems, 2014(12): 37-41.)
[15] 石菊, 王小倩, 王熙. 微观医疗费用预测模型：从线性回归到机器学习[J]. 经济学( 季刊), 2023, 23(6): 2247-2263. (SHI J, WANG X Q, WANG X. Modeling individual medical expenditures: From linear regression to machine learning [J]. China economic quarterly, 2023, 23(6): 2247-2263.)
[16] 雷可为, 陈瑛. 基于BP神经网络和ARIMA组合模型的中国入境游客量预测[J]. 旅游学刊, 2007(4): 20-25. (LEI K W, CHEN Y. Forecast of inbound tourists to China based on BP neural network and ARIMA combined model [J]. Tourism tribune, 2007(4): 20-25.)
[17] 王向前, 吴东隆, 郑健彤. 货物吞吐量预测的改进ARIMAX方法——以天津港为例[J]. 运筹与管理, 2022, 31(3): 138- 144. (WANG X Q, WU D L, ZHENG J T. Improved ARIMAX method for cargo throughput forecasting: a case study of Tianjin port [J]. Operations research and management science, 2022, 31(3): 138-144.)
[18] 洪永淼, 汪寿阳. 大数据如何改变经济学研究范式？ [J]. 管理 世界, 2021, 37(10): 40-55, 72, 56. (HONG Y M, WANG S Y. How is big data changing economic research paradigms?[J]. Journal of management world, 2021, 37(10): 40-55, 72, 56.)
[19] 李云良, 张奇, 李淼, 等. 基于BP神经网络的鄱阳湖水位模拟[J]. 长江流域资源与环境, 2015, 24(2): 233-240. (LI Y L, ZHANG Q, LI M, et al. Using BP neural networks for water level simulation in Poyang Lake[J]. Resources and environment in the Yangtze Basin, 2015, 24(2): 233-240.)
[20] 杨青, 王晨蔚. 基于深度学习LSTM神经网络的全球股票指数预测研究[J]. 统计研究, 2019, 36(3): 65-77. (YANG Q, WANG C W. A study on forecast of global stock indices based on deep LSTM neural network [J]. Statistical research, 2019, 36(3): 65-77.)
[21] 王艳艳, 蔡钟文, 于李胜. 基于机器学习的商誉减值预测模型 [J]. 会计研究, 2024(3): 51-64. (WANG Y Y, CAI Z W, YU L S. Predicting goodwill impairment using machine learning [J]. Accounting research, 2024(3): 51-64.)
[22] XIE G, QIAN Y, WANG S, et al. A decomposition-ensemble approach for tourism forecasting[J]. Annals of tourism research, 2020, 81: 102891.
[23] LI X, ZHANG X, ZHANG C, et al. Forecasting tourism demand with a novel robust decomposition and ensemble framework[J]. Expert systems with applications, 2024, 236: 121388.
[24] 唐振鹏, 吴俊传, 张婷婷, 等. 基于二次分解和集成学习的粮食期货价格预测研究[J]. 系统工程理论与实践, 2021, 41(11): 2837-2849. (TANG Z P, WU J C, ZHANG T T, et al. Research on grain fiutures price forecasting based on secondary decomposition and ensemble learning [J]. Systems engineeringtheory & practice, 2021, 41(11): 2837-2849.)
[25] LI M, CHENG Z, LIN W, et al. What can be learned from the historical trend of crude oil prices? An ensemble approach for crude oil price forecasting[J]. Energy economics, 2023, 123: 106736.
[26] 高晓辉, 周坤, 李廉水. 基于XGBOOST和ELM的混合空气质量预警系统：以南京为例[J]. 中国管理科学, 2023, 31(5): 269-278. (GAO X H, ZHOU K, LI L S. Hybrid air quality early warning system based on XGBoost and ELM: A case study of Nanjing [J]. Chinese journal of management science, 2023, 31(5): 269-278.)
[27] DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE transactions on signal processing, 2013, 62(3): 531-544.
[28] WANG J, LI X, HONG T, et al. A semi-heterogeneous approach to combining crude oil price forecasts[J]. Information sciences, 2018, 460: 279-292.
[29] 梁小珍, 郭战坤, 张倩文, 等. 基于奇异谱分析的航空客运需求分析与分解集成预测模型[J]. 系统工程理论与实践, 2020, 40(7): 1844-1855. (LIANG X Z, GUO Z K, ZHANG Q W, et al. An analysis and decomposition ensemble prediction model for air passenger demand based on singular spectrum analysis [J]. Systems engineering-theory& practice, 2020, 40(7): 1844-1855.)
[30] 陈龙, 刘刚, 戚聿东, 等. 人工智能技术革命：演进、影响和 应对 [J]. 国际 经济 评论, 2024(3): 9-51, 4. (CHEN L, LIU G, QI Y D, et al. AI technology revolution: Evolution, impact and countermeasures [J]. International economic review, 2024(3): 9-51, 4.)
[31] 熊凯军. 经济政策不确定性、企业异质性与技术创新——基于我国上市制造业企业经验分析[J]. 软科学, 2021, 35(6): 15-22. (XIONG K J. Economic policy uncertainty, enterprise heterogeneity and technological innovation: based on the experience analysis of my country’s listed manufacturing companies [J]. Soft science, 2021, 35(6): 15-22.)
[32] 张海波, 戴新宇, 钱德沛, 等. 新一代信息技术赋能应急管理现代化的战略分析[J]. 中国科学院院刊, 2022, 37(12): 1727-1737. (ZHANG H B, DAI X Y, QIAN D P, et al. Strategic perspective of leveraging new generation information technology to enable modernization of emergency management [J]. Bulletin of Chinese Academy of Sciences, 2022, 37(12): 1727-1737.)
Options
文章导航

/

〈 〉