Measuring knowledge exploration distance at the patent level: Application of network embedding and citation analysis |
| |
| Institution: | 1. Computational Science Research Center, Korea Institute of Science and Technology, Seoul, Republic of Korea;2. Department of Industrial Engineering, Konkuk University, Seoul, Republic of Korea;2. Department of Industrial Engineering, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 08826, Republic of Korea;1. Warsaw University of Technology, Faculty of Mathematics and Information Science, ul. Koszykowa 75, Warsaw 00-662, Poland;2. Deakin University, School of IT, Geelong, VIC 3220, Australia;3. Systems Research Institute, Polish Academy of Sciences, ul. Newelska 6, Warsaw 01-447, Poland;4. Warsaw University of Technology, Faculty of Physics, ul. Koszykowa 75, Warsaw 00-662, Poland;1. Department of Statistics, Cochin University of Science and Technology, Cochin 682022, India;2. Department of Statistics, Government Arts College, Thiruvananthapuram 695014, India;1. Institute of Scientific and Technical Information of China, Beijing 100038, P.R. China;2. College of Economics and Management, Beijing University of Technology, Beijing 100124, P.R. China;3. School of Information Resource Management, Renmin University of China, Beijing 100872, P.R. China;4. Business School, Shandong University of Technology, Zibo 255000, P.R. China;1. Department of Library and Information Science Education, College of Education, Kongju National University, Gongju 32588, Republic of Korea;2. Department of Library and Information Science, Hannam University, Daejeon 34430, Republic of Korea;3. Department of Library and Information Science, Cheongju University, Cheongju 28503, Republic of Korea;4. Department of Library and Information Science, Yonsei University, Seoul 03722, Republic of Korea |
| |
| Abstract: | Although the incidence of knowledge exploration is observed in most patents, the concept of knowledge exploration distance has been analyzed with limited patents at the macro level of a company, organization or region. This study quantifies the knowledge exploration distance of individual patents using network embedding methods and citation analysis. First, a technology ecology network is constructed to identify technological association relationships between technical elements. Second, network embedding method is employed to represent technical elements as fixed dimensional vector, preserving the structural information. Next, the individual patents are vectorized based on the technology classification code information and pre-trained embedding values. Finally, by comparing the position between a citing patent and cited patents in the vector space, the knowledge exploration distance of the patent is obtained. This knowledge exploration distance indicates the novel degree of technological association between technical elements of a citing patent and those of cited patents. The case study covering artificial intelligence technology-related patents is conducted to illustrate the process of calculating knowledge exploration distance. Besides, this study showed that the proposed measure has significant relationships with patent-based indicators related to protection coverage, prior knowledge, and patent value. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
