RESEARCH PAPERS

Predicting Patent Tradability Using Cat Boost: An Empirical Study Based on U.S. Physics Patents from 2002 to 2012

  • Ding Qunzhe ,
  • Song Yonghui ,
  • Song Hongjie ,
  • Ran Congjing
Expand
  • 1 School of Information Management, Wuhan University, Wuhan 430072;
    2 Center for Studies of Information Resources, Wuhan University, Wuhan 430072
Ding Qunzhe, doctoral candidiate; Song Yonghui, doctoral candidiate, corresponding author, E-mail: yonghuisong@whu.edu.cn.;Song Hongjie, master’s candidiate; Ran Congjing, professor, PhD, doctoral supervisor.

Received date: 2024-09-09

  Revised date: 2025-02-13

  Online published: 2025-08-01

Supported by

This work is supported by the general project of National Natural Science Foundation of China, titled“High-Quality Patent Identification and Evolution Analysis in Emerging Technology Fields Using Graph Convolutional Neural Networks” (Grant No. 72274084).

Abstract

[Purpose/Significance] Identifying and predicting patent tradability is vital for implementing the innovation-driven strategy, accelerating technology commercialization, and promoting industry–research collaboration. [Method/Process] This paper treated patents as technological units and screened tradable ones based on prior research. Using the Incopat and YiPatent databases, patent transactions and their related technical characteristics were analyzed through statistical methods. Based on this, a prediction model for tradable patents was proposed, incorporating multidimensional features. The task was framed as a supervised binary classification, utilizing the legal, technical, and economic features of patents to predict the likelihood of a transaction before expiration, with machine learning models trained accordingly. [Result/Conclusion] The results are as follows. Patents with higher tradability are characterized by fewer patent holders, a greater number of citations, and more claims. The priority filling gap is the most significant indicator of patent tradability, with longer gaps reducing the likelihood of transactions. The CatBoost(Categorical Boosting) model demonstrates strong overall performance and training efficiency, making it the most effective model for predicting patent tradability. Nonetheless, predicting patent transactions remains challenging, and future research could improve performance by incorporating patent text and image features.

Cite this article

Ding Qunzhe , Song Yonghui , Song Hongjie , Ran Congjing . Predicting Patent Tradability Using Cat Boost: An Empirical Study Based on U.S. Physics Patents from 2002 to 2012[J]. Library and Information Service, 2025 , 69(16) : 79 -89 . DOI: 10.13266/j.issn.0252-3116.2025.16.007

References

[1] 雷聪仪, 张娴.科研院校技术转移效率评价研究进展述评与展望[J]. 图书情报工作, 2022, 66(19):132-142.(LEI C Y, ZHANG X. Review and prospects of research on efficiency evaluation of technology transfer in scientific research institutions[J]. Library and information service, 2022, 66(19):132-142.)
[2] DELERUE H. Shadow of joint patents:intellectual property rights sharing by smes in contractual R & D alliances[J]. Journal of business research, 2018, 87:12-23.
[3] KIM M, GEUM Y. Predicting patent transactions using patentbased machine learning techniques[J]. IEEE access, 2020, 8:188833-188843.
[4] LICHTENTHALER U. Open innovation in practice:an analysis of strategic approaches to technology transactions[J]. IEEE transactions on engineering management, 2008, 55(1):148-157.
[5] YOON J, PARK H, SEO W, et al. Technology opportunity discovery(TOD)from existing technologies and products:a function-based TOD framework[J]. Technological forecasting and social change, 2015, 100:153-167.
[6] SEO W, YOON J, PARK H, et al. Product opportunity identification based on internal capabilities using text mining and association rule mining[J]. Technological forecasting and social change, 2016, 105:94-104.
[7] YOON J, SEO W, COH B Y, et al. Identifying product opportunities using collaborative filtering-based patent analysis[J]. Computers & industrial engineering, 2017, 107:376-387.
[8] 张娴, 胡正银, 茹丽洁, 等.专利技术供需信息关联知识组织模式研究[J]. 图书情报工作, 2016, 60(8):118-125.(ZHANG X, HU Z Y, RU L J, et al. Research on the knowledge organization model of patent technology supply-demand information association[J]. Library and information service, 2016, 60(8):118-125.)
[9] 吴逸菲, 樊春良.创新系统视角下美国国家人工智能战略的演化逻辑及趋势分析[J]. 科学学与科学技术管理, 2024, 45(7):29-48.(WU Y F, FAN C L. Evolution logic and trend analysis of the U.S. national artificial intelligence strategy from the perspective of innovation systems[J]. Science of science and management of s. & t., 2024, 45(7):29-48.)
[10] 国务院.国务院办公厅关于印发《专利转化运用专项行动方案(2023-2025年)》的通知[EB/OL]. [2025-06-17]. https://www.gov.cn/gongbao/2023/issue_10806/202311/content_6913817.html.(The State Council. The General Office of the State Council issued the Notice on the Special Action Plan for Patent Transformation and Utilization(2023-2025)[EB/OL]. [2025-06-17]. https://www.gov.cn/gongbao/2023/issue_10806/202311/content_6913817.html.)
[11] 国家知识产权局. 2022年中国专利调查报告[EB/OL]. [2025-06-30]. https://www.cnipa.gov.cn/module/download/down.jsp?i_ID=181043& colID=88.(China National Intellectual Property Administration(CNIPA). 2022 China Patent Survey Report[EB/OL]. [2025-06-30]. https://www.cnipa.gov.cn/module/download/down.jsp?i_ID=181043& colID=88.)
[12] DE MARCO A, SCELLATO G, UGHETTO E, et al. Global markets for technology:evidence from patent transactions[J]. Research policy, 2017, 46(9):1644-1654.
[13] MONK A H B. The emerging market for intellectual property:drivers, restrainers, and implications[J]. Journal of economic geography, 2009, 9(4):469-491.
[14] DENTER N M, SEEGER F, MOEHRLE M G. How can blockchain technology support patent management? a systematic literature review[J]. International journal of information management, 2023, 68:102506.
[15] HUANG H C, SHIH H Y, KE T H. Strategic intentions of patent transactions:a network analysis of the tft-lcd industry[J]. IEEE transactions on engineering management, 2022, 71(5):2063-2076.
[16] ANANTHRAMAN S, CAMBRÉB, KITTLER M, et al. Divide and conquer:relating patent quality and value in a conceptual framework based on a systematic review[J]. International journal of management reviews, 2024, 26(2):285-311.
[17] BURHOP C, WOLF N. The German market for patents during the“second industrialization, “1884-1913:a gravity approach[J]. Business history review, 2013, 87(1):69-93.
[18] CHEN Z R, NIE P Y, HE C, et al. The impacts of auction methods and market structure on patent portfolio value[J]. Technology analysis & strategic management, 2022, 34(11):1295-1309.
[19] 何喜军, 董艳波, 武玉英, 等.基于异构信息网络嵌入的专利技术主体间交易推荐模型[J]. 情报学报, 2020, 39(1):57-67.(HE X J, DONG Y B, WU Y Y, et al. Patent transaction recommendation model between technology entities based on heterogeneous information network embedding[J]. Journal of the China Society for Scientific and Technical Information, 2020, 39(1):57-67.)
[20] 何喜军, 吴爽爽, 武玉英, 等.基于属性异构网络表示学习的专利交易推荐[J]. 情报学报, 2022, 41(11):1214-1228.(HE X J, WU S S, WU Y Y, et al. Patent transaction recommendation based on attribute heterogeneous network representation learning[J]. Journal of the China Society for Scientific and Technical Information, 2022, 41(11):1214-1228.)
[21] 何喜军, 石安杰, 吴爽爽, 等.基于知识图谱与强化学习的专利交易推荐研究[J]. 系统工程理论与实践, 2024, 44(10):3330-3345.(HE X J, SHI A J, WU S S, et al. Research on patent transaction recommendation based on knowledge graph and reinforcement learning[J]. Systems engineering:theory & practice, 2024, 44(10):3330–3345.)
[22] CHOI J, JANG D, JUN S, et al. A predictive model of technology transfer using patent analysis[J]. Sustainability, 2015, 7(12):16175-16195.
[23] QI X L, ZHAO C P, GONG Y, et al. Predicting future new technology trade relations among regional comprehensive economic partnership member countries-based on a link prediction method[J]. International journal of technology management, 2023, 93(2):133-164.
[24] 何喜军, 孟雪, 武玉英, 等.基于二分网络多维投影的技术交易机会预测——以新能源领域专利技术为例[J]. 管理评论, 2022, 34(7):105-114.(HE X J, MENG X, WU Y Y, et al. Prediction of technology transaction opportunities based on multidimensional projection of bipartite networks:a case study of patent technology in the new energy field[J]. Management review, 2022, 34(7):105-114.)
[25] 胡锴, 朱雪忠, 王义, 等.网上技术交易趋势预测模型及应用研究[J]. 科技进步与对策, 2020, 37(22):1-8.(HU K, ZHU X Z, WANG Y, et al. Research on online technology transaction trend prediction models and their applications[J]. Science & technology progress and policy, 2020, 37(22):1-8.)
[26] 武玉英, 才久然, 何喜军.基于多层关系网络的专利技术主体间交易机会预测——以电子信息技术领域为例[J]. 情报杂志, 2020, 39(7):49-53, 100.(WU Y Y, CAI J R, HE X J.Prediction of patent technology transaction opportunities among entities based on multilayer relational networks:a case study in the electronic information technology field[J]. Journal of intelligence, 2020, 39(7):49-53, 100.)
[27] JUN S, PARK S, JANG D. A technology valuation model using quantitative patent analysis:a case study of technology transfer in big data marketing[J]. Emerging markets finance and trade, 2015, 51(5):963-974.
[28] SANDRIK K. Proving priority:assigning employee inventions in a global market[J]. University of Pennsylvania journal of international law, 2020, 41(3):755-808.
[29] ZHANG Y, QIAN Y, HUANG Y, et al. An entropy-based indicator system for measuring the potential of patents in technological innovation:rejecting moderation[J]. Scientometrics, 2017, 111(3):1925-1946.
[30] 国家市场监督管理总局, 国家标准化管理委员会.专利评估指引:GB/T 42748-2023[S]. 北京:中国标准出版社, 2023.(The State Administration for Market Regulation, the Standardization Administration of China. Guidelines for Patent Evaluation:GB/T42748-2023[S]. Beijing:Standards Press of China, 2023.)
[31] 白英晨.专利价值评价指标体系研究[J]. 经济研究导刊, 2019(10):21-23, 117.(BAI Y C. Research on the indicator system of patent value evaluation[J]. Economic research guide, 2019(10):21-23, 117.)
[32] KO N, JEONG B, SEO W, YOON J. A transferability evaluation model for intellectual property[J]. Computers & industrial engineering, 2019, 131:344-355.
[33] 李欣, 冯野, 马晓迪.一种基于机器学习的专利可转让性评估方法研究[J]. 情报杂志, 2023, 42(3):85-93.(LI X, FENG Y, MA X D. Research on a patent assignability evaluation method based on machine learning[J]. Journal of intelligence, 2023, 42(3):85-93.)
[34] 武玉英, 才久然, 何喜军.结合高阶神经元的深度神经网络用于专利可转让性评价——以电子信息技术领域为例[J]. 情报杂志, 2021, 40(2):63-68.(WU Y Y, CAI J R, HE X J. Deep neural networks with high-order neurons for patent assignability evaluation:a case study in the electronic information technology field[J]. Journal of intelligence, 2021, 40(2):63-68.)
[35] 杨冠灿, 刘彤, 李纲, 等.基于综合引用网络的专利价值评价研究[J]. 情报学报, 2014, 32(12):1265-1277.(YANG G C, LIU T, LI G, et al. Research on patent value evaluation based on an integrated citation network[J]. Journal of the China Society for Scientific and Technical Information, 2014, 32(12):1265-1277.)
[36] VIMALNATH P, GURTOO A , MATHE W M. Patent characteristics and the age-value relationship:study of oceantomo auctioned US singleton patents for the period 2006-2008[J]. R & D management, 2017, 47(5):738-750.
[37] 张杰, 孙超, 翟东升, 等.基于诉讼专利的专利质量评价方法研究[J]. 科研管理, 2018, 39(5):138-146.(ZHANG J, SUN C, ZHAI D S, et al. Research on patent quality evaluation methods based on litigation patents[J]. Science research management, 2018, 39(5):138-146.)
[38] SU F P, LAI K K, SHARMA R R, et al. Patent priority network:linking patent portfolio to strategic goals[J]. Journal of the American Society for Information Science and Technology, 2009, 60(11):2353-2361.
[39] GRIMALDI M, CRICELLI L, ROGO F. Valuating and analyzing the patent portfolio:the patent portfolio value index[J]. European journal of innovation management, 2018, 21(2):174-205.
[40] 王子焉, 刘文涛, 倪渊, 等.专利价值评估研究综述[J]. 科技管理研究, 2019, 39(16):181-190.(WANG Z Y, LIU W T, NI Y, et al. A review of patent value evaluation research[J]. Science and technology management research, 2019, 39(16):181-190.)
[41] DUNN M K. Timing of patent filing and market exclusivity[J]. Nature reviews drug discovery, 2011, 10(7):487-488.
[42] RASSENFOSSE G, DERNIS H, GUELLEC D, et al. The worldwide count of priority patents:a new indicator of inventive activity[J]. Research policy, 2013, 42(3):720-737.
[43] 王子焉, 倪渊, 张健.基于灰色关联分析-随机森林回归的网络平台专利价值评估方法研究[J]. 情报理论与实践, 2019, 42(10):109-116.(WANG Z Y, NI Y, ZHANG J. Research on network platform patent value evaluation methods based on grey relational analysis and random forest regression[J]. Information theory and practice, 2019, 42(10):109-116.)
[44] GAO L, PORTER A, WANG J, et al. Technology life cycle analysis method based on patent documents[J]. Technological forecasting and social change, 2013, 80(3):398-407.
[45] 金花, 刘玥, 张更平.单项专利价值的评估与定量评估指标体系的构建——基于邻域粗糙集与果蝇优化神经网络的单项专利价值评估[J]. 大学图书馆学报, 2020, 38(3):48-56, 64.(JIN H, LIU Y, ZHANG G P. Evaluation of individual patent value and construction of a quantitative evaluation index system:an individual patent value evaluation based on neighborhood rough sets and fruit fly optimization neural network[J]. Journal of academic libraries, 2020, 38(3):48-56, 64.)
[46] 孙玉涛, 栾倩.专利质量测度“三阶段-两维度”模型及实证研究——以C9联盟高校为例[J]. 科学学与科学技术管理, 2016, 37(6):23-32.(SUN Y T, LUAN Q. Patent quality measurement“three-stage-two-dimension” model and empirical study:a case study of C9 alliance universities[J]. Science of science and management of s. & t., 2016, 37(6):23-32.)
[47] TANHA J, ABDI Y, SAMADI N, et al. Boosting methods for multi-class imbalanced data classification:an experimental review[J]. Journal of big data, 2020, 7(1):70.
[48] LUNDBERG S M, ERION G, CHEN H, et al. From local explanations to global understanding with explainable AI for trees[J]. Nature machine intelligence, 2020, 2(1):56-67.
[49] BOWERS A J, ZHOU X. Receiver operating characteristic(roc)area under the curve(auc):a diagnostic measure for evaluating the accuracy of predictors of education outcomes[J]. Journal of education for students placed at risk, 2019, 24(1):20-46.
Outlines

/

〈 〉