An entropy-based indicator system for measuring the potential of patents in technological innovation: rejecting moderation

Publication Type:
Journal Article
Citation:
Scientometrics, 2017, pp. 1 - 22
Issue Date:
2017-03-07
Full metadata record
Files in This Item:
Filename Description Size
offprint - SCIM GTM SI Revision.pdfAccepted Manuscript Version583.2 kB
Adobe PDF
© 2017 Akadémiai Kiadó, Budapest, HungaryHow to evaluate the value of a patent in technological innovation quantitatively and systematically challenges bibliometrics. Traditional indicator systems and weighting approaches mostly lead to “moderation” results; that is, patents ranked to a top list can have only good-looking values on all indicators rather than distinctive performances in certain individual indicators. Orienting patents authorized by the United States Patent and Trademark Office (USPTO), this paper constructs an entropy-based indicator system to measure their potential in technological innovation. Shannon’s entropy is introduced to quantitatively weight indicators and a collaborative filtering technique is used to iteratively remove negative patents. What remains is a small set of positive patents with potential in technological innovation as the output. A case study with 28,509 USPTO-authorized patents with Chinese assignees, covering the period from 1976 to 2014, demonstrates the feasibility and reliability of this method.
Please use this identifier to cite or link to this item: