摘要识别潜在颠覆性技术有助于国家和企业加强颠覆性技术供给，使其在国际科技竞争中赢得竞争优势或实现变轨超车。传统基于文献计量的颠覆性技术识别方法通常利用论文和专利数据先构建关键词网络或关键词集，然后人工构造高阶数据特征进行分析。这种人工构造高阶特征的方法容易使关键词网络等的结构信息表达不充分，导致识别的准确性降低。本文提出一种基于图表示学习的半监督潜在颠覆性技术识别方法。首先，基于科技文献数据库数据，利用关键词共现频率和期刊影响力构建技术关键词加权网络；然后，通过反向传播算法基于匿名游走序列学习获得关键词网络的向量表示；接着，通过比较待识别技术关键词网络的向量序列与公认颠覆性技术关键词网络的向量序列之间的相似程度，反映技术演化特征的相似性，从而识别出潜在的颠覆性技术；最后，从近年来国内外与颠覆性技术有关的战略规划、预测报告中选取10项技术作为实验对象，采集WoS（Web of Science）数据对本文提出的方法进行实验验证，发现在预给定5项颠覆性技术的条件下，本文方法能较好地将其中潜在的3项颠覆性技术识别出来，并能够将2项伪颠覆性技术判断为非颠覆性技术。
1 开庆, 窦永香. 颠覆性技术识别研究综述[J]. 情报杂志, 2021, 40(11): 31-38. 2 张金柱, 张晓林. 利用引用科学知识突变识别突破性创新[J]. 情报学报, 2014, 33(3): 259-266. 3 白光祖, 郑玉荣, 吴新年, 等. 基于文献知识关联的颠覆性技术预见方法研究与实证[J]. 情报杂志, 2017, 36(9): 38-44. 4 孙永福, 王礼恒, 孙棕檀, 等. 引发产业变革的颠覆性技术内涵与遴选研究[J]. 中国工程科学, 2017, 19(5): 9-16. 5 Blume M, Oberl?nder A M, R?glinger M, et al. Ex ante assessment of disruptive threats: identifying relevant threats before one is disrupted[J]. Technological Forecasting and Social Change, 2020, 158: 120103. 6 Dixon T, Eames M, Britnell J, et al. Urban retrofitting: identifying disruptive and sustaining technologies using performative and foresight techniques[J]. Technological Forecasting and Social Change, 2014, 89: 131-144. 7 Ganguly A, Nilchiani R, Farr J V. Defining a set of metrics to evaluate the potential disruptiveness of a technology[J]. Engineering Management Journal, 2010, 22(1): 34-44. 8 Drew S A W. Building technology foresight: using scenarios to embrace innovation[J]. European Journal of Innovation Management, 2006, 9(3): 241-257. 9 Burt G. Why are we surprised at surprises? Integrating disruption theory and system analysis with the scenario methodology to help identify disruptions and discontinuities[J]. Technological Forecasting and Social Change, 2007, 74(6): 731-749. 10 Brimley S, Fitzgerald B, Sayler K. Game changers: disruptive technology and U.S. defense strategy[R]. Washington D.C.: Center for a New American Security, 2013. 11 Linton J D. Forecasting the market diffusion of disruptive and discontinuous innovation[J]. IEEE Transactions on Engineering Management, 2002, 49(4): 365-374. 12 Adner R. When are technologies disruptive? A demand-based view of the emergence of competition[J]. Strategic Management Journal, 2002, 23(8): 667-688. 13 Chen C, Zhang J, Guo R S. The D-Day, V-Day, and bleak days of a disruptive technology: a new model for ex-ante evaluation of the timing of technology disruption[J]. European Journal of Operational Research, 2016, 251(2): 562-574. 14 Stelzer B, Meyer-Br?tz F, Schiebel E, et al. Combining the scenario technique with bibliometrics for technology foresight: the case of personalized medicine[J]. Technological Forecasting and Social Change, 2015, 98: 137-156. 15 Momeni A, Rost K. Identification and monitoring of possible disruptive technologies by patent-development paths and topic modeling[J]. Technological Forecasting and Social Change, 2016, 104: 16-29. 16 Cheng Y, Huang L C, Ramlogan R, et al. Forecasting of potential impacts of disruptive technology in promising technological areas: Elaborating the SIRS epidemic model in RFID technology[J]. Technological Forecasting and Social Change, 2017, 117: 170-183. 17 黄鲁成, 成雨, 吴菲菲, 等. 关于颠覆性技术识别框架的探索[J]. 科学学研究, 2015, 33(5): 654-664. 18 Bloodworth I. A search for discriminative linguistic markers in ICT practitioner discourse, for the ex ante identification of disruptive innovation[D]. Wellington: Victoria University of Wellington, 2012. 19 Dotsika F, Watkins A. Identifying potentially disruptive trends by means of keyword network analysis[J]. Technological Forecasting and Social Change, 2017, 119: 114-127. 20 Li X, Xie Q Q, Jiang J J, et al. Identifying and monitoring the development trends of emerging technologies using patent analysis and Twitter data mining: the case of perovskite solar cell technology[J]. Technological Forecasting and Social Change, 2019, 146: 687-705. 21 李乾瑞, 郭俊芳, 黄颖, 等. 基于突变-融合视角的颠覆性技术主题演化研究[J]. 科学学研究, 2021, 39(12): 2129-2139. 22 Chen X L, Han T. Disruptive technology forecasting based on gartner hype cycle[C]// Proceedings of the 2019 IEEE Technology & Engineering Management Conference. Los Angeles: IEEE, 2019: 1-6. 23 Zhang D K, Yin J, Zhu X Q, et al. Network representation learning: a survey[J]. IEEE Transactions on Big Data, 2020, 6(1): 3-28. 24 Shi B X, Yang J, Weninger T, et al. Representation learning in heterogeneous professional social networks with ambiguous social connections[C]// Proceedings of the 2019 IEEE International Conference on Big Data. Los Angeles: IEEE, 2019: 1928-1937. 25 Li D Y, Lin Q, Ma X K. Identification of dynamic community in temporal network via joint learning graph representation and nonnegative matrix factorization[J]. Neurocomputing, 2021, 435: 77-90. 26 Ge S Y, Wu C H, Wu F Z, et al. Graph enhanced representation learning for news recommendation[C]// Proceedings of the Web Conference 2020. New York: ACM Press, 2020: 2863-2869. 27 余传明, 林奥琛, 钟韵辞, 等. 基于网络表示学习的科研合作推荐研究[J]. 情报学报, 2019, 38(5): 500-511. 28 张鑫, 文奕, 许海云. 一种融合表示学习与主题表征的作者合作预测模型[J]. 数据分析与知识发现, 2021, 5(3): 88-100. 29 Nikzad-Khasmakhi N, Balafar M A, Feizi-Derakhshi M R, et al. BERTERS: multimodal representation learning for expert recommendation system with transformers and graph embeddings[J]. Chaos, Solitons & Fractals, 2021, 151: 111260. 30 宁原隆, 周刚, 卢记仓, 等. 一种融合关系路径与实体描述信息的知识图谱表示学习方法[J]. 计算机研究与发展, 2022, 59(9): 1966-1979. 31 陈鑫, 刘喜恩, 吴及. 药物表示学习研究进展[J]. 清华大学学报(自然科学版), 2020, 60(2): 171-180. 32 Fout A, Byrd J, Shariat B, et al. Protein interface prediction using graph convolutional networks[C]// Proceedings of the Advances in Neural Information Processing Systems. Cambridge: MIT Press, 2017: 6530-6539. 33 Gilmer J, Schoenholz S S, Riley P F, et al. Neural message passing for quantum chemistry[C]// Proceedings of the 34th International Conference on Machine Learning. New York: ACM Press, 2017: 1263-1272. 34 孙亚伟, 程龚, 厉肖, 等. 基于图匹配网络的可解释知识图谱复杂问答方法[J]. 计算机研究与发展, 2021, 58(12): 2673-2683. 35 Saxena A, Tripathi A, Talukdar P. Improving multi-hop question answering over knowledge graphs using knowledge base embeddings[C]// Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2020: 4498-4507. 36 王一华. 基于IF (JCR)、IF (Scopus)、H指数、SJR值、SNIP值的期刊评价研究[J]. 图书情报工作, 2011, 55(16): 144-148. 37 Micali S, Zhu Z A. Reconstructing Markov processes from independent and anonymous experiments[J]. Discrete Applied Mathematics, 2016, 200: 108-122. 38 Lau J H, Baldwin T. An empirical evaluation of doc2vec with practical insights into document embedding generation[C]// Proceedings of the 1st Workshop on Representation Learning for NLP. Stroudsburg: Association for Computational Linguistics, 2016: 78-86. 39 Yang Z C, Yang D Y, Dyer C, et al. Hierarchical attention networks for document classification[C]// Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: Association for Computational Linguistics, 2016: 1480-1489. 40 Mikolov T, Chen K, Corrado G, et al. Efficient estimation of word representations in vector space[OL]. (2013-09-07). https://arxiv.org/abs/1301.3781. 41 李海林, 邬先利. 基于时间序列聚类的主题发现与演化分析研究[J]. 情报学报, 2019, 38(10): 1041-1050. 42 Shokoohi-Yekta M, Hu B, Jin H X, et al. Generalizing DTW to the multi-dimensional case requires an adaptive approach[J]. Data Mining and Knowledge Discovery, 2017, 31(1): 1-31. 43 王康, 陈悦, 宋超, 等. 颠覆性技术: 概念辨析与特征分析[J]. 科学学研究, 2022, 40(11): 1937-1946. 44 赵国屏. 合成生物学: 开启生命科学“会聚”研究新时代[J]. 中国科学院院刊, 2018, 33(11): 1135-1149. 45 陈云伟, 陶诚, 周海晨, 等. 基因编辑技术研究进展与挑战[J]. 世界科技研究与发展, 2021, 43(1): 8-23.
