Text matching to measure patent similarity
Сопоставление текстов для измерения сходства патентов
2017-09-14
SCID: 54.1/2wkdknyq
Discuss with AI
knowledge spilloverspatent portfoliospatent similaritytechnological similaritytext matching
Figures from the paper
Abstract (AI)
Research Summary : We propose using text matching to measure the technological similarity between patents. Technology experts from different fields validate the new similarity measure and its improvement on measures based on the United States Patent Classification System, and identify its limitations. As an application, we replicate prior findings on the localization of knowledge spillovers by constructing a case–control group of text‐matched patents. We also provide open access to the code and data to calculate the similarity between any two utility patents granted by the United States Patent and Trademark Office between 1976 and 2013, or between any two patent portfolios. Managerial Summary : We propose using text matching to measure the technological similarity between patents. The method can be used by various practitioners such as inventors, attorneys, patent examiners, and managers to search for closely related prior art, to assess the novelty of a patent, to identify R&D opportunities in less crowded areas, to detect in‐ or out‐licensing opportunities, to map companies in technology space, and to find acquisition targets. We use an expert panel to validate the improvement of the new similarity measure on measures based on the United States Patent Classification System, and provide open access to the code and data to calculate the similarity between any two utility patents granted by the USPTO between 1976 and 2013, or between any two patent portfolios.
Key Findings
1
Technology experts validate that the text-based similarity measure improves upon measures based on the United States Patent Classification System.
2
The authors provide open-access code and data for calculating similarity between USPTO utility patents granted from 1976 to 2013 and between patent portfolios.
3
The method replicates prior findings on geographically localized knowledge spillovers using text-matched patent case–control groups.
4
The method supports practical tasks including prior-art search, patent novelty assessment, R&D opportunity identification, licensing analysis, technology mapping, and acquisition-target discovery.
5
The paper introduces a text-matching method for measuring technological similarity between patents.
Research Object
utility patents and patent portfolios granted by the United States Patent and Trademark Office between 1976 and 2013
Research Subject
technological similarity between patents and the validity, improvement, and limitations of text-matching-based similarity measurement
Publication Details
Publication Date
2017-09-14
Journal
Publisher
ISSN
Cited by
268
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest