图书情报工作

图书情报工作 >

2025 , Vol. 69 >Issue 13: 19 - 31

DOI: https://doi.org/10.13266/j.issn.0252-3116.2025.13.002

研究论文

融合多重专利关系网络指标的高质量专利识别研究

  • 邓伟伟 ,
  • 霍栩俊 ,
  • 吴敏 ,
  • 陈寒 ,
  • 奉国和
展开
  • 1 华南师范大学经济与管理学院 广州 510006;
    2 华南师范大学教师教育学部 广州 510631
邓伟伟，副教授，博士，硕士生导师；霍栩俊，硕士研究生；吴敏，本科生；陈寒，副研究员，博士，硕士生导师，通信作者，E-mail:1107162944@qq.com；奉国和，教授，博士，硕士生导师。

收稿日期: 2024-09-05

  修回日期: 2024-11-04

  网络出版日期: 2025-06-21

基金资助

本文系2024年度国家社会科学基金项目“多方法融合视角下高价值专利挖掘及影响因素识别研究”（项目编号：24BTQ036）研究成果之一。

收起

Research on High-Quality Patents Identification by Integrating Network Indicators from Multiple Patent Relationships

  • Deng Weiwei ,
  • Huo Xujun ,
  • Wu Min ,
  • Chen Han ,
  • Feng Guohe
Expand
  • 1 School of Economics and Management, South China Normal University, Guangzhou 510006;
    2 College of Teacher Education, South China Normal University, Guangzhou 510631
Deng Weiwei, associate professor, PhD, master supervisor; Huo Xujun, master candidate; Wu Min, undergraduate student; Chen Han, associate researcher, PhD, master supervisor, corresponding author, E-mail: 1107162944@qq.com; Feng Guohe, professor, PhD, master supervisor.

Received date: 2024-09-05

  Revised date: 2024-11-04

  Online published: 2025-06-21

Supported by

This work is supported by the project of National Social Science Fund of China, titiled “Research on High-Value Patent Mining and Influencing Factor Identification from the Perspective of Multi-Method Integration” (Grant No. 24BTQ036).

Fold

摘要

[目的/意义] 考虑专利引用网络、发明网络、类别关联网络中节点结构重要性，提出融合多重专利关系网络指标的高质量专利识别方法。[方法/过程] 基于专利的引用、发明人和分类信息构建多重专利关系网络，从中提取度中心度、中介中心度、接近中心度、特征向量中心度和PageRank指标衡量专利节点结构重要性，然后融合网络指标与传统专利质量评估指标形成指标体系，最后将其输入到机器学习模型以判断专利是否高质量。[结果/结论] 基于美国专利商标局数据的实证分析表明，融合多重专利关系网络指标能够显著提升高质量专利识别效果，且模型特征重要性排序前4位中有两个网络指标，这进一步说明网络指标在高质量专利识别中的关键作用。

关键词： 高质量专利; 网络特征; 专利识别; 机器学习

本文引用格式

邓伟伟 , 霍栩俊 , 吴敏 , 陈寒 , 奉国和 . 融合多重专利关系网络指标的高质量专利识别研究[J]. 图书情报工作, 2025 , 69(13) : 19 -31 . DOI: 10.13266/j.issn.0252-3116.2025.13.002

Abstract

[Purpose/Significance] Considering the importance of node structure in patent citation networks, invention networks, and category association networks, this research proposes a novel method for identifying high-quality patents by integrating multiple indicators from patent relationship networks. [Method/Process] Based on patent citation, inventor, and classification information, a network with multiple patent relationships was constructed. Degree centrality, betweenness centrality, closeness centrality, eigenvector centrality, and PageRank indicators were extracted from the network to measure the structural importance of patent nodes. These network indicators were then integrated with traditional patent quality evaluation indicators to form an indicator system, which was subsequently input into machine learning models to determine whether a patent was of high quality. [Result/Conclusion] Empirical analysis based on data from the United States Patent and Trademark Office demonstrates that the integration of multiple patent relationship network indicators significantly improves the identification of high-quality patents. Furthermore, two of the top four ranked important features are network indicators, further emphasizing the critical role of network indicators in identifying high-quality patents.

Key words： high-quality patent; network feature; patent identification; machine learning

参考文献

[1] HIGHAM K, RASSENFOSSE G D, JAFFE A. Patent quality: towards a systematic framework for analysis and measurement[J]. Research policy, 2021, 50(4): 104215.
[2] 刘立春. 二元专利质量研究[J]. 情报杂志, 2017, 36(11): 168-174. (LIU L C. Research on binary patent quality[J]. Journal of intelligence, 2017, 36(11): 168-174.)
[3] 彭华涛, 田兰馨. 高质量专利评估研究进展及展望[J]. 武汉理工大学学报(社会科学版), 2022, 35(3): 48-56. (PENG H T, TIAN L X. Progress and prospect of high quality patent evaluation[J]. Journal of Wuhan University of Technology(social sciences edition), 2022, 35(3): 48-56.)
[4] 朱雪忠, 万小丽. 竞争力视角下的专利质量界定[J]. 知识产权, 2009, 19(4): 7-14. (ZHU X Z, WANG X L. The definition of patent quality from a competitiveness perspective[J]. Intellectual property, 2009, 19(4): 7-14.)
[5] 许鑫, 赵文华, 姚占雷. 多维视角的高质量专利识别及其应用研究[J]. 现代情报, 2019, 39(11): 13-22, 45. (XU X, ZHAO W H, YAO Z L. Research on the identification and application of high-quality patents: from the perspective of multiple dimensions[J]. Journal of modern information, 2019, 39(11): 13-22, 45.)
[6] 吴洁, 桂亮, 刘鹏. 基于图卷积网络的高质量专利自动识别方案研究[J]. 情报杂志, 2022, 41(1): 88-95, 124. (WU J, GUI L, LIU P. Indicator and textual features-based patent evaluation with graph convolutional networks[J]. Journal of intelligence, 2022, 41(1): 88-95, 124.)
[7] SCHANKERMAN M, PAKES A. Estimates of the value of patent rights in European countries during the post-1950 period[J]. Economic journal, 1986, 96(384): 1052-1076.
[8] DE RASSENFOSSE G, JAFFE A B. Are patent fees effective at weeding out low-quality patents[J]. Journal of economics and management strategy, 2018, 27(1): 134-148.
[9] 周一夫, 谭春辉, 江婷, 等.基于机器学习分类算法的高质量专利成果筛选研究[J]. 现代情报, 2024, 44(2): 81-91. (ZHOU Y F, TAN C H, JIANG T, et al. Research on the screening method of high-quality patent[J]. Journal of modern information, 2024, 44(2): 81-91.)
[10] 曹勇, 王子欣, 肖丽. 基于文献计量的专利质量研究现状、合作网络特征与演进趋势分析[J]. 创新科技, 2023, 23(7): 66-78. (CAO Y, WANG Z X, XIAO L. An analysis of status, cooperation networks and evolutionary trends of patent quality research based on bibliometrics[J]. Innovation science and technology, 2023, 23(7): 66-78.)
[11] 尹承锐, 孙世强, 陈明明.专利高质量发展趋势及其技术创新驱动机制研究[J]. 情报杂志, 2022, 41(4): 70-76, 97. (YIN C R, SUN S Q, CHEN M M. The high-quality patent development trend and driving mechanism of technological innovation[J]. Journal of intelligence, 2022, 41(4): 70-76, 97.)
[12] 郭青, 戚湧, 高盼军. 基于技术、法律和经济三位一体的专利质量评价及应用研究[J]. 中国发明与专利, 2021, 18(1): 21-29. (GUO Q, QI Y, GAO P J. Research on evaluation and application of patent quality based on the trinity model of technology quality, legal validity and economic perspective[J]. China invention & patent, 2021, 18(1): 21-29.)
[13] 冯君, 周静珍, 杜芸. 单件专利质量评价指标体系研究[J]. 科技管理研究, 2012, 32(23): 166-170. (FENG J, ZHOU J Z, DU Y. Study on the quality evaluation method for single patent[J]. Science and technology management research, 2012, 32(23): 166-170.)
[14] 李欣, 范明姐, 黄鲁成. 基于机器学习的专利质量评价研究[J]. 科技进步与对策, 2020, 37(24): 116-124. (LI X, FAN M J, HUANG L C. Research on patent quality evaluation using machine learning[J]. Science & technology progress and policy, 2020, 37(24): 116-124.)
[15] 谢祥, 马晓雅, 卢彤, 等. 基于GRU-Attention机制的高校发明专利质量评估模型研究[J]. 科技管理研究, 2021, 41:79-85. (XIE X, MA X Y, LU T, et al. Research on quality evaluation model of university invention patent based on GRU-Attention mechanism[J]. Science and technology management research, 2021, 41(7): 79-85.)
[16] CHUNG P, SOHN S Y. Early detection of valuable patents using a deep learning model: Case of semiconductor industry[J]. Technological forecasting and social change, 2020, 158:120146.
[17] SUH J H. Exploring the effect of structural patent indicators in forward patent citation networks on patent price from firm market value[J]. Technology analysis & strategic management, 2015, 27(5): 485-502.
[18] BEAUDRY C, SCHIFFAUEROVA A. Impacts of collaboration and network indicators on patent quality: the case of Canadian nanotechnology innovation[J]. European management journal, 2011, 29(5): 362-376.
[19] KWON U, GEUM Y. Identification of promising inventions considering the quality of knowledge accumulation: a machine learning approach[J]. Scientometrics, 2020, 125(3): 1877-1897.
[20] 李黎明, 张敏, 李小娟. 引证网络专利质量对专利拍卖经济价值的影响效应研究[J]. 情报杂志, 2021, 40(10): 115-121. (LI L M, ZHANG M, LI X J. Effect of the quality of patent of citation network on patent economic value at auction[J]. Journal of intelligence, 2021, 40(10): 115-121.)
[21] 刘雯, 曹思未, 叶静怡. 社会网络与高校专利技术成果转移[J]. 世界经济, 2020, 43(9): 173-192. (LIU M, CAO S W, YE J Y. Social networks and university technology transfer[J]. The journal of world economy, 2020, 43(9): 173-192.)
[22] 孙冰, 徐晓菲, 苏晓. 技术扩散主路径及核心企业的识别研究——以手机芯片专利引文网络为例[J]. 情报学报, 2019, 38(2): 201-208. (SUN B, XU X F, SU X. Identification of the main path of technology diffusion and core enterprises: a case study of patent citation network in mobile phone chips[J]. Journal of the China Society for Scientific and Technical Information, 2019, 38(2): 201-208.)
[23] HUANG W R, HSIEH C J, CHANG K C, et al. Network characteristics and patent value: evidence from the light-emitting diode industry[J]. Plos one, 2017, 12(8): e0181988.
[24] 冉从敬, 李旺, 胡启彪, 等. 基于机器学习的成本法在专利价值评估中的应用研究——以“新能源汽车”为例[J]. 现代情报, 2024, 44(5): 140-152. (RAN C H, LI W, HU Q B, et al. Research on the application of machine learning-based cost method in patent value assessment: taking “New Energy Vehicle”as the case[J]. Journal of modern information, 2024, 44(5): 140-152.)
[25] 王思培, 韩涛. 基于随机森林算法的潜在高价值专利预测方法研究[J]. 情报科学, 2020, 38(5): 120-125. (WANG S P, HAN T. Prediction method of potential high-value patents based on random forest algorithm[J]. Information science, 2020, 38(5): 120-125.)
[26] 张耀天, 杜慰纯, 贾明顺, 等. 基于自适应层次分析法的企业专利质量评价研究[J]. 图书情报工作, 2016, 60(7): 110-115. (ZHANG Y T, DU W C, JIA M S, et al. Research on the evaluation of enterprise patents quality based on the adaptive analytic hierarchy process[J]. Library and information service, 2016, 60(7): 110-115.)
[27] 施晴, 王芸, 徐宏, 等. 基于专利转化的高校生物医药专利质量评价研究[J]. 科技管理研究, 2019, 39(11): 139-145. (SHI Q, WANG Y, XU H, et al. Quality evaluation of biomedical patents based on colleges and universities patent transformation[J]. Science and technology management research, 2019, 39(11): 139-145.)
[28] 孙冉, 安璐, 李纲.多特征融合的专利价值预测——以5G技术为例[J]. 现代情报, 2022, 42(11): 87-96. (SUN R, AN L, LI G. Patent value prediction based on multi-feature fusion: taking the 5G technology as a case[J]. Journal of modern information, 2022, 42(11): 87-96.)
[29] 刘兰剑, 张萌, 黄天航. 政府补贴、税收优惠对专利质量的影响及其门槛效应——基于新能源汽车产业上市公司的实证分析[J]. 科研管理, 2021, 42(6): 9-16. (LIU L J, ZHANG M, HUANG T H. The impact of government subsidies and tax preferences on patent quality and its threshold effect: an empirical analysis based on the listed companies in the new energy automobile industry[J]. Science research management, 2021, 42(6): 9-16.)
[30] 周成, 魏红芹. 专利价值评估与分类研究——基于自组织映射支持向量机[J]. 数据分析与知识发现, 2019, 3(5): 117-124. (ZHOU C, WEI H Q. Evaluating and classifying patent values based on self-organizing maps and support vector machine[J]. Data analysis and knowledge discovery, 2019, 3(5): 117-124.)
[31] YANG Y, XIE G. Efficient identification of node importance in social networks[J]. Information Processing and Management, 2016, 52(5): 911-922.
[32] 罗浩, 闫光辉, 张萌, 等. 融合多元信息的多关系社交网络节点重要性研究[J]. 计算机研究与发展, 2020, 57(5): 954-970. (LUO H, YAN G H, ZHANG M, et al. Research on node importance fused multi-information for multi-relational social networks[J]. Journal of computer research and development, 2020, 57(5): 954-970.)
[33] ANDREAS M, UTE J F, MALTE D, et al. Interactive effects of social network centrality and social identification on stress[J]. British journal of psychology, 2020, 112(1): 144-162.
[34] DEY P, CHATERJEE A, ROY S. Influence maximization in online social network using different centrality measures as seed node of information propagation[J]. Sādhanā, 2019, 44(9): 1-13.
[35] 陈玉, 胡泽文, 周西姬. 国际高价值专利：概念内涵、主题范畴、多维价值指标与评估预测方法综述[J]. 现代情报, 2024, 44(8): 153-170. (CHEN Y, HU Z W, ZHOU X J. Overview of international high value patents: conceptual connotation, theme categories, multidimensional values, and evaluation and prediction methods[J]. Journal of modern information, 2024, 44(8): 153-170.)
[36] 张宪立, 唐建新. 一种新的复杂网络节点重要性评估方法[J]. 计算机工程, 2021, 47(2): 139-145, 151. (ZHANG X L, TANG J X.A novel evaluation method of node importance in complex network[J]. Computer engineering, 2021, 47(2): 139-145, 151.)
[37] 朱丽波. 科学合作网络的中心性分析[J]. 图书馆学研究, 2015, (3): 97-101. (ZHU L B. Academic social network centrality analysis[J]. Research on library science, 2015, (3): 97-101.)
[38] 张克群, 项星星, 张婷, 等. 识别高被引专利——基于稀有事件Logit与倾向得分匹配模型[J]. 图书馆论坛, 2021, 41(6): 67-74. (ZHANG K Q, XIANG X X, ZHANG T, et al. Identifying highly cited patents: based on rare event logit and propensity score matching model[J]. library tribune, 2021, 41(6): 67-74.)
Options
文章导航

/

〈 〉