[目的/意义] 探索融合引用和文本特征的专利技术创新路径识别分析方法，有助于规避技术创新风险、优化选择技术创新路径，对提升创新主体的创新能力，促进现代产业发展，布局科技前沿发展战略等具有重要的意义。[方法/过程] 首先基于 Node2Vec 模型和 Doc2Vec 模型将专利引用和文本数据表示学习为可计算的高维向量；然后利用 LDA 主题模型进行技术主题识别并结合 T-SNE 算法降维，添加时间维度构建初始技术创新路径；最后，在专利引用和文本特征向量表示结果基础上，开展向量融合拼接从而实现融合引用和文本特征的技术创新路径识别。[结果/结论] 通过对超级电容器领域的实证，验证提出的融合引用和文本特征的的技术创新路径识别方法能够从特定领域专利文献中高效、准确地识别专利技术创新路径，证明方法的可行性和有效性。
[Purpose/Significance] Exploring the identification and analysis method of patent technology innovation path integrating citation and text features is helpful to avoid the risk of technology innovation, optimize the selection of technology innovation path, and is of great significance to improve the innovation ability of innovation subjects, promote the development of modern industry, and lay out the development strategy of science and technology frontier. [Method/Process] Firstly, based on node2vec model and doc2vec model, the patent reference and text data representation were learned as computable high-dimensional vectors, and then the LDA topic model was used to identify the technical topic and the t-sne algorithm was used to reduce the dimension, and the time dimension was added to build the initial technical innovation path. Finally, vector stitching and fusion were carried out to realize the technology innovation path recognition integrating references and text features based on the vector representation results of patent reference and text features. [Result/Conclusion] Through the demonstration in the field of supercapacitors, it is verified that the technology innovation path identification method proposed in this paper, which integrates citation and text features, can identify the patent technology innovation path efficiently and accurately from the patent documents in specific fields, and the feasibility and effectiveness of the method proposed in this paper are verified.
[1] "十三五"国家科技创新规划[EB/OL].[2022-10-17].http://www.gov.cn/xinwen/2016-08/08/content_5098259.htm.
[2] 王金凤, 徐正强, 冯立杰, 等.基于多维空间专利地图及可拓学的技术创新路径识别与评价[J].科技管理研究, 2022, 42(8):8-17.
[3] 颜端武, 白敬毅, 李晨晨, 等.科技领域技术演进的多路径识别及创新特性分析[J].情报理论与实践, 2021, 44(11):99-107.
[4] 冯立杰, 尤鸿宇, 王金凤.专利技术创新路径识别及其新颖性评价研究[J].情报学报, 2021, 40(5):513-522.
[5] WISSEMA J G.Morphological analysis:its application to a company TF investigation[J].Futures, 1976, 8(2):146-153.
[6] ILEVBARE I M, PROBERT D, PHAAL R.A review of TRIZ, and its benefits and challenges in practice[J].Technovation, 2013, 33(2/3):30-37.
[7] MITCHELL V W.The Delphi technique:an exposition and application[J].Technology analysis & strategic management, 1991, 3(4):333-358.
[8] VERSPAGEN B.Mapping technological trajectories as patent citation networks:a study on the history of fuel cell research[J].Advances in complex systems, 2007, 10(1):93-115.
[9] MARTINELLI A.An emerging paradigm or just another trajectory? understanding the nature of technological changes using engineering heuristics in the telecommunications switching industry[J].Research policy, 2012, 41(2):414-429.
[10] 严素梅, 吉久明, 陈荣, 等.多维度创新路径识别与发现研究[J].图书馆杂志, 2020, 39(9):104-110.
[11] 宋歌.共被引分析方法迭代创新路径研究[J].情报学报, 2020, 39(1):12-24.
[12] JAMES T, COOK D F, CONLON S, et al.A framework to explore innovation at SAP through bibliometric analysis of patent applications[J].Expert systems with application, 2015, 42(24):9389-9401.
[13] 周潇, 黄璐, 马婷婷.大数据视角下的技术创新路径识别研究[J].科研管理, 2017, 38(10):1-9.
[14] 王雅薇, 周源, 陈璐怡.我国人工智能产业技术创新路径识别及分析——基于专利分析法[J].科技管理研究, 2019, 39(10):210-216.
[15] 白如江, 孙一钢, 张庆芝.基于知识基因表达的科技创新路径构建研究[J].情报理论与实践, 2020, 43(4):137-144, 176.
[16] 冯立杰, 尤鸿宇, 王金凤.专利技术创新路径识别及其新颖性评价研究[J].情报学报, 2021, 40(5):513-522.
[17] 杨雯雯.基于创新基因理论的技术创新路径应用研究[D].郑州:郑州大学, 2021.
[18] 王金凤, 徐正强, 冯立杰, 等.基于多维空间专利地图及可拓学的技术创新路径识别与评价[J].科技管理研究, 2022, 42(8):8-17.
[19] 曹安业, 刘耀琪, 杨旭, 等.物理指标与数据特征融合驱动的冲击地压时序预测方法[J/OL].煤炭学报:1-16[2022-07-19].https://kns.cnki.net/kcms2/article/abstract?v=3uoqIhG8C45S0n9fL2suRadTyEVl2pW9UrhTDCdPD67H3i82VzEfE-P-AwNjBaYeNkYxFwoJXhxXgrIaD4BnHK8o_2oZQCYC&uniplatform=NZKPT.
[20] 檀莹莹, 王俊丽, 张超波.基于图卷积神经网络的文本分类方法研究综述[J].计算机科学, 2022, 49(8):205-216.
[21] 葛轶洲, 刘恒, 王言, 等.小样本困境下的深度学习图像识别综述[J].软件学报, 2022, 33(1):193-210.
[22] GROVER A, LESKOVEC J.node2vec:Scalable feature learning for networks[C]//Proceeding of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining.New York:ACM, 2016:855-864.
[23] LE Q V, MIKOLOV T.Distributed representations of sentences and documents[J].JMLR.org, 2014(5):1188-1196.
[24] BLEI D M, NG A Y, JORDAN M I.Latent Dirichlet allocation[J].Journal of machine learning research, 2003, 3(1):993-1022.
[25] BLEI D M, LAFFERTY J.Dynamic topic models[C]//Proceedings of the 23rd international conference on machine learning.New York:ACM, 2006:113-120.
[26] 刘自强, 许海云, 岳丽欣, 等.面向研究前沿预测的主题扩散演化滞后效应研究[J].情报学报, 2018, 37(10):979-988.
[27] LAURENS V D M, HINTON G.Visualizing data using t-SNE[J].Journal of machine learning research, 2008, 9(11):2579-2605.
