Structural decomposition of technological domain using patent co-classification and classification hierarchy
- Authors
- Mun, Changbae; Yoon, Sejun; Park, Hyunseok
- Issue Date
- Nov-2019
- Publisher
- SPRINGER
- Keywords
- Classification overlap method (COM); Patent co-classification; Classification hierarchy; Sub-technologies; Sub-domain; Hierarchical class similarity
- Citation
- SCIENTOMETRICS, v.121, no.2, pp.633 - 652
- Indexed
- SCIE
SSCI
SCOPUS
- Journal Title
- SCIENTOMETRICS
- Volume
- 121
- Number
- 2
- Start Page
- 633
- End Page
- 652
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/146827
- DOI
- 10.1007/s11192-019-03223-8
- ISSN
- 0138-9130
- Abstract
- This paper proposes a new method for decomposing a technological domain (TD). Specifically, the method identifies sub-TDs at the different levels of technological hierarchy within the TD based on the characteristics of patent co-classification and classification hierarchy. We defined the smallest class, named Minimum Overlapped Class (MOC), constructed by overlaps of sub-group IPC(s) and sub-class UPC(s), and sub-TD is basically identified as a set of the MOCs. In order to cluster the MOCs, technological distances among MOCs are calculated based on patent co-classification and hierarchical structure of patent classification systems. Technologically similar MOCs are grouped by using a hierarchical clustering and the identified clusters at the different level of hierarchy show the hierarchical structure of a TD. Detailed technological content for each sub-TD is represented by extracting representative keywords through a text-mining technique. The method is empirically tested by the solar photovoltaic technology and the results show that the identified sub-TDs are reasonably acceptable by qualitative analysis.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 정보시스템학과 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.