Consistency and Trends of Technological Innovations: A Network Approach to the International Patent Classification Data

Yuan Gao, Zhen Zhu, Massimo Riccaboni

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)
8 Downloads (Pure)

Abstract

Classifying patents by the technology areas they pertain is important to enable information search and facilitate policy analysis and socio-economic studies. Based on the OECD Triadic Patent Family database, this study constructs a cohort network based on the grouping of IPC subclasses in the same patent families, and a citation network based on citations between subclasses of patent families citing each other. This paper presents a systematic analysis approach which obtains naturally formed network clusters identified using a Lumped Markov Chain method, extracts community keys traceable over time, and investigates two important community characteristics: consistency and changing trends. The results are verified against several other methods, including a recent research measuring patent text similarity. The proposed method contributes to the literature a network-based approach to study the endogenous community properties of an exogenously devised classification system. The application of this method may improve accuracy and efficiency of the IPC search platform and help detect the emergence of new technologies.
Original languageEnglish
Title of host publicationInternational Conference on Complex Networks and their Applications
Subtitle of host publicationCOMPLEX NETWORKS 2017: Complex Networks & Their Applications VI
PublisherSpringer International Publishing AG
Pages744-756
VolumeVI
ISBN (Electronic)978-3-319-72150-7
ISBN (Print)978-3-319-72149-1
DOIs
Publication statusPublished - Nov 2017

Cite this