基于专利相似度的突破性技术发明识别研究——以纳米技术为例

摘要/Abstract

摘要： 随着我国发明专利申请数量的迅猛增加，如何通过事前和事后指标测度并识别技术和经济价值高的突破性技术发明就成为学术界面临的焦点问题。针对我国专利普遍缺乏引文信息的现状，本文利用专利的国际专利分类（IPC）信息构建两两专利相似度指标，并引入时间维度对过去、当前以及未来三个时间段的专利相似度比较，测度专利的新颖性、独特性和影响力，从而构建突破性技术发明的综合识别方案。然后，以纳米技术为例，利用美国专利商标局（USPTO）在1975-2015年的授权发明专利数据进行实证检验。结果表明：（1）基于专利IPC四位和六位分类的相似度指标分别可以识别出6.23%和5.06%的纳米技术专利为突破性技术发明；（2）基于专利相似度识别的突破性技术发明与基于专利被引数识别的突破性技术发明具有显著的正相关关系，但是，两类识别方法得到的结果中仅有不足总样本的0.5%是相同的，表明以往单纯依赖专利被引数据识别突破性技术发明可能存在一定偏差；（3）对突破性技术发明来源特征的实证检验表明，基于专利相似度和基于专利被引数的突破性技术发明的发明人和组织来源特征基本一致，而发明层面的知识来源特征呈现不一致的结果，进一步反映出两类识别方案的差异性。本文基于专利相似度构建的突破性技术发明识别方案既为企业在实践中挖掘和利用高价值的发明专利提供参考，也对未来突破性技术发明相关研究达成一致结论具有重要意义。

关键词: 专利相似度, IPC分类, 突破性技术发明, 识别方案, 纳米技术

Abstract: With the rapid increase in the number of patent applications in China, how to measure and identify breakthrough inventions with high technology and economic value has become a hot issue. Since 2011, the number of invention patent applications in China has ranked first in the world for eight consecutive years. However, how to identify breakthrough inventions through ex ante and ex post indicators remains a problem in academia. Thus, it is of significance to identify the invention patents with major breakthroughs and turn them into productivity in China.
Considering the general lack of citation information in Chinese patents, this paper draws on the definitions and identification criteria of breakthrough inventions by Dahlin and Behrens (2005) to construct the patent-to-patent similarity index with the IPC information of patents. Considering both the ex ante and ex post indicators of patents, we introduces the time dimension to measure the novelty, uniqueness, and impact of patents by comparing the similarity of patents in the past, current and future periods, so proposing a comprehensive identification solution for breakthrough inventions. In order to test the validity of our identification solution, we choose the field of nanotechnology as an example. As one of the fastest growing and widely influential emerging science and technology fields in the world, nanotechnology is a typical interdisciplinary technology field. The cross-domain attributes of nanotechnology determine the wide range of IPC classifications involved, which is suitable for our research using patent IPC classification and is more useful to reveal the invention characteristics of technologies in the emerging interdisciplinary domains. Then, we conduct an empirical test by using the granted invention patents in nanotechnology from the United States Patent and Trademark Office (USPTO) in 1989-2010.
Our conclusions are as follows. First, we find that it is feasible and operational to use IPC classification similarity to identify breakthrough inventions. Specifically, we can identify 6.23% and 5.06% of nanotechnology patents as breakthrough inventions with the similarity indicators based on IPC four- and six-digit classifications. Second, by comparing with the breakthrough inventions identified based on patent citation, this paper verifies the validation and difference of the breakthrough inventions identified based on patent IPC classification similarity. This paper finds that there is a significant positive correlation between breakthrough inventions identified based on patent similarity and based on patent forward citations. At the same time, however, this paper finds that only 0.5% of the results obtained by the two types of identification methods are the same, indicating that it may be biased by purely relying on patented citation data to identify breakthrough inventions in the past, so the identification solution based on IPC classification similarity is a useful supplement. Finally, through the comparative study of the source characteristics at the invention-, inventor- and organizational-level, this paper further validates the effectiveness and difference of the breakthrough inventions identified based on patent IPC classification similarity. The breakthrough inventions based on patent similarity and patent citation shows basically consistent results in inventor and organizational characteristics, indicating the validation of the new identification solution based on patent IPC similarity in the inventor- and organizational-level. At the same time, however, this paper also finds that the two types of identification methods show inconsistent results in the characteristics of the invention′s knowledge source, indicating that the previous research results on the source characteristics of the invention may be affected by the breakthrough invention identification method. Thus, we suggest that future research on the source of breakthrough inventions needs to pay attention to different identification methods.

Key words: patent similarity, IPC classification, breakthrough invention, identification method, nanotechnology

马荣康王艺棠. 基于专利相似度的突破性技术发明识别研究——以纳米技术为例[J]. 科研管理, 2021, 42(5): 153-160.

Ma Rongkang, Wang Yitang. Identification of breakthrough inventions based on patent-to-patent similarity: A case study of nanotechnology[J]. Science Research Management, 2021, 42(5): 153-160.

参考文献

[1] 陈傲,柳卸林. 突破性技术从何而来?——一个文献评述[J]. 科学学研究, 2011, 29(9):1281-1290.

Chen Ao, Liu Xielin. How are radical technology developed?——A literature review[J]. Studies in Science of Science, 2011, 29(9):1281-1290.

[2] 陈劲, 戴凌燕, 李良德. 突破性创新及其识别[J]. 科技管理研究, 2002, 22(5): 22-28.

Chen Jin, Dai Lingyan, Li Liangde. Radical innovation and its identification[J]. Science and Technology Management Research, 2002, 22(5): 22-28.

[3] Verhoeven D, Bakker J, Veugelers R. Measuring technological novelty with patent-based indicators[J]. Research Policy, 2016, 45(3): 707-723.

[4] Ahuja G, Morris Lampert C. Entrepreneurship in the large corporation: A longitudinal study of how established firms create breakthrough inventions[J]. Strategic management journal, 2001, 22(6-7): 521-543.

[5] Phene A, Fladmoe‐Lindquist K, Marsh L. Breakthrough innovations in the US biotechnology industry: the effects of technological space and geographic origin[J]. Strategic Management Journal, 2006, 27(4): 369-388.

[6] Zheng Y, Yang H. Does familiarity foster innovation? The impact of alliance partner repeatedness on breakthrough innovations[J]. Journal of Management Studies, 2015, 52(2): 213-230.

[7] Schoenmakers W, Duysters G. The technological origins of radical inventions[J]. Research Policy, 2010, 39(8): 1051-1059.

[8] Conti R, Gambardella A, Mariani M. Learning to be Edison: inventors, organizations, and breakthrough inventions[J]. Organization Science, 2013, 25(3): 833-849.

[9] Singh J, Fleming L. Lone inventors as sources of breakthroughs: Myth or reality?[J]. Management science, 2010, 56(1): 41-56.

[10] Datta A, Jessup L M. Looking beyond the focal industry and existing technologies for radical innovations[J]. Technovation, 2013, 33(10): 355-367.

[11] Qi Dong J, McCarthy K J, Schoenmakers W W M E. How central is too central? Organizing interorganizational collaboration networks for breakthrough innovation[J]. Journal of Product Innovation Management, 2017, 34(4): 526-542.

[12] Dahlin K B, Behrens D M. When is an invention really radical?: Defining and measuring technological radicalness[J]. Research Policy, 2005, 34(5): 717-737.

[13] 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.

Zhang Jinzhu, Zhang Xiaolin. Identification of Radical Innovation Based on Mutation of Cited Scientific Knowledge [J]. Journal of the China Society for Scientific and Technical Information, 2014, 33(3): 259-266.

[14] 张金柱, 张晓林. 基于专利科学引文的突破性创新识别研究述评[J]. 情报学报, 2016, 35(9): 955-962.

Zhang Jinzhu, Zhang Xiaolin. Overview of Radical Innovation Identification Based on Scientific References in Patents [J]. Journal of the China Society for Scientific and Technical Information, 2016, 35(9): 955-962.

[15] Kaplan S, Vakili K. The double‐edged sword of recombination in breakthrough innovation[J]. Strategic Management Journal, 2015, 36(10): 1435-1457.

[16] 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.

Huang Lucheng, Cheng Yu, Wu Feifei, et al. Study on identification framework of disruptive innovation[J]. Studies in Science of Science, 2015, 33(5): 654-664.

[17] Arts S, Cassiman B, Gomez J C. Text matching to measure patent similarity[J]. Strategic Management Journal, 2018, 39(1): 62-84.

[18] Rodriguez A, Kim B, Turkoz M, et al. New multi-stage similarity measure for calculation of pairwise patent similarity in a patent citation network[J]. Scientometrics, 2015, 103(2): 565-581.

[19] McNamee R C. Can’t see the forest for the leaves: Similarity and distance measures for hierarchical taxonomies with a patent classification example[J]. Research Policy, 2013, 42(4): 855-873.

[20] Aharonson B S, Schilling M A. Mapping the technological landscape: Measuring technology distance, technological footprints, and technology evolution[J]. Research Policy, 2016, 45(1): 81-96.

[21] 彭继东, 谭宗颖. 一种基于文本挖掘的专利相似度测量方法及其应用[J]. 情报理论与实践, 2010, 33(12): 114-118.

Peng Jidong, Tan Zongying. A Method and Its Application of Patent Similarity Measurement Based on Text Mining [J]. Information studies: Theory & Application, 2010, 33(12): 114-118.

[22] 吴菲菲, 李倩, 黄鲁成. 基于专利 SAO 结构的技术应用领域识别方法研究[J]. 科研管理, 2014, 35(6): 1-7.

Wu Feifei, Li Qian, Huang Lucheng. The method of identifying the application field of technology based on the SAO structure of patents [J]. Science Research Management, 2014, 35(6): 1-7.

[23] Strumsky D, Lobo J, Van der Leeuw S. Using patent technology codes to study technological change[J]. Economics of Innovation and New technology, 2012, 21(3): 267-286.

[24] OECD. Nanotechnology: an overview based on indicators and statistics[W]. OECD Science, Technology and Industry Working Papers 2009/07, Directorate for Science, Technology and Industry, 2009.

[25] Wong P. K., Ho Y. P., Chan C. K. Internationalization and evolution of application areas of an emerging technology: The case of nanotechnology[J]. Scientometrics, 2007, 70(3): 715-737.

[26] Hill C W L, Rothaermel F T. The performance of incumbent firms in the face of radical technological innovation[J]. Academy of Management Review, 2003, 28(2): 257-274.