A Product Patent Knowledge Mining and Recommendation Method Based on Few-shot Feature Extraction and Cross-citation Network

doi:10.3901/JME.260100

Abstract

Abstract: To address the challenges of insufficient knowledge mining and low efficiency of knowledge reuse in the innovative design of complex industrial products, a complex product patent knowledge mining and recommendation method based on few-shot feature extraction and cross-citation network is proposed. Firstly, patent knowledge model is constructed by extracting three types of featured entities including domain(D), function(F), and technology(T), which is represented as a three-aspect model <D, F, T>. Meanwhile, the annotation rule of few-shot patent dataset is proposed for manual feature annotation. Secondly, a fine-tuned BERT model is constructed based on a few-shot annotated patent dataset to achieve automatic patent feature extraction. A graph neural network is employed to learn the representation of the patent co-citation network. Thirdly, the fine-tuned BERT model is subsequently used to extract feature entities in the input design problem or task description, which are vectorized and their cosine similarities computed to generate a recommendation list. Finally, a prototype based on Django framework is also developed, which realizes the two functions of feature extraction and knowledge push, and illustrates the use scenarios of this method based on a case of automobile windshield wiper strip design.

Key words: innovative design, patent recommendation, patent knowledge mining, named entity recognition

CLC Number:

TG156

WANG Zuoxu, LI Mingrui, ZHANG Muye, LIU Jihong. A Product Patent Knowledge Mining and Recommendation Method Based on Few-shot Feature Extraction and Cross-citation Network[J]. Journal of Mechanical Engineering, 2026, 62(3): 492-504.

References

[1] 王克勤，毋凤君. 面向产品设计改进的在线评论挖掘[J]. 计算机工程与应用，2019，55(19)：235-245，252. WANG Keqin，WU Fengjun. Online review mining for product design improvement[J]. Computer Engineering and Applications，2019，55(19)：235-245，252.
[2] 孔德婧，董放，陈子婧，等. 离群专利视角下的新兴技术预测——基于BERT模型和深度神经网络[J]. 图书情报工作，2021，65(17)：131-141. KONG Dejing，DONG Fang，CHEN Zijing，et al. Emerging technology forecasting from the perspective of outlier patents：Based on BERT model and deep neural network[J]. Library and Information Service，2021，65(17)：131-141.
[3] 祝德刚，宫琳，唐圣，等. 基于专利知识图谱的产品创新概念设计方法[J]. 计算机集成制造系统，2022，28(11)：3599-3614. ZHU Degang，GONG Lin，TANG Sheng，et al. Conceptual design method for product innovation based on patent knowledge graph[J]. Computer Integrated Manufacturing Systems，2022，28(11)：3599-3614.
[4] ZHAI D，ZHAI L，LI M，et al. Patent representation learning with a novel design of patent ontology：Case study on PEM patents[J]. Technological Forecasting and Social Change，2022，183：121912.
[5] PARK H，KIM K，CHOI S，et al. A patent intelligence system for strategic technology planning[J]. Expert Systems with Applications，2013，40(7)：2373-2390.
[6] 吴正. 机电产品专利知识挖掘技术和应用研究[D]. 贵阳：贵州大学，2017. WU Zheng. Research on patent knowledge mining technology and application for electromechanical products[D]. Guiyang：Guizhou University，2017.
[7] 王嘉瑜. 基于知识图谱的专利自动推荐技术研究[D]. 哈尔滨：哈尔滨工业大学，2021. WANG Jiayu. Research on patent automatic recommendation technology based on knowledge graph[D]. Harbin：Harbin Institute of Technology，2021.
[8] 郭鑫，赵武，王杰，等. 面向创新设计的工艺设计知识模型及检索方法研究[J]. 机械工程学报，2017，53(15)：66-72. GUO Xin，ZHAO Wu，WANG Jie，et al. Research on process design knowledge model and retrieval method for innovative design[J]. Journal of Mechanical Engineering，2017，53(15)：66-72.
[9] 董文斌，战洪飞，余军合，等. 机械产品专利知识的提取和应用[J]. 机械制造，2021，59(8)：1-8. DONG Wenbin，ZHAN Hongfei，YU Junhe，et al. Extraction and application of patent knowledge in mechanical products[J]. Machinery Manufacturing，2021，59(8)：1-8.
[10] SARICA S，LUO J，WOOD K L. TechNet：Technology semantic network based on patent data[J]. Expert Systems with Applications，2020，142：112995.
[11] 李少波，吴正. 基于类比设计的专利知识挖掘算法[J]. 计算机集成制造系统，2017，23(11)：2344. LI Shaobo，WU Zheng. Patent knowledge mining algorithm based on analogical design[J]. Computer Integrated Manufacturing Systems，2017，23(11)：2344.
[12] 林园园，战洪飞，余军合，等. 数据驱动的产品概念设计知识服务模型构建[J]. 计算机工程与应用，2018，54(16)：211-219. LIN Yuanyuan，ZHAN Hongfei，YU Junhe，et al. Data-driven knowledge service model construction for product concept design[J]. Computer Engineering and Applications，2018，54(16)：211-219.
[13] LIU W，LIU J，WU M，et al. Representation learning over multiple knowledge graphs for knowledge graphs alignment[J]. Neurocomputing，2018，320：12-24.
[14] 余秀. 基于深度学习的专利知识推送方法研究[D]. 杭州：浙江大学，2020. YU Xiu. Research on patent knowledge push method based on deep learning[D]. Hangzhou：Zhejiang University，2020.
[15] KIM J，YOON J，PARK E，et al. Patent document clustering with deep embeddings[J]. Scientometrics，2020，123(2)：563-577.
[16] LEI L，QI J，ZHENG K. Patent analytics based on feature vector space model：A case of IoT[J]. IEEE Access，2019，7：45705-45715.
[17] 吴秀丽，马隆洲，向东，等. 面向退役机电产品全生命周期的知识图谱统一建模方法研究[J]. 机械工程学报，2003，59(7)：1-16. WU Xiuli，MA Longzhou，XIANG Dong，et al. Unified modeling method of knowledge graph for the full life cycle of retired electromechanical products[J]. Journal of Mechanical Engineering，2003，59(7)：1-16.
[18] TRAPPEY A J C，LIANG C P，LIN H J. Using machine learning language models to generate innovation knowledge graphs for patent mining[J]. Applied Sciences，2022，12(19)：9818.