Download database  Download article
IPRoduct project Prof. Gaétan de Rassenfosse Dr. David Portabella Dissemination Software and data Restricted Contact Us

IPRoduct: Linking Patents to Products

IPRoduct logo

About the Authors

Gaétan de Rassenfosse is Assistant Professor at the College of Management of Technology (CDM) at Ecole polytechnique fédérale de Lausanne (EPFL). David Portabella is a software engineer at the CDM.


The project seeks to link innovative goods sold in the U.S. economy to the patents upon which they are based. By directly linking patents to products, this project tracks innovation to the point where it meets consumers, the true commercial end point of investments in science & technology. This project provides a new tool for demonstrating in a more systematic manner the breadth of science, the role of federal funding, and their combined impact on everyday life.

The construction of the database exploits the virtual patent marking (VPM) statute introduced in the 2011 Leahy-Smith America Invents Act. Under that act, patentees may give notice to the public that their product is patented. The project creates software that is able to identify VPM sections of websites and to parse the product-patent links to populate the database.

The output of the project is a pilot database of linked patent-product pairs that will be made publicly available. The production of such a database has never been attempted on such a large scale before.


We gratefully acknowledge financial support from the U.S. NSF (award number 1645264) and the European Patent Office. We are also grateful to CommonCrawl, EPFL-Scitas and CSCS teams for providing access to their infrastructure and support.

Complete Article

An article is in preparation and will be available soon.

Access the Data

The data will be made publicly available upon completion of the project.


The IPRoduct project was presented at various workshops and conferences including:

"Notice failure revisited: Evidence on the use of virtual patent marking",
Gaétan de Rassenfosse, EFPL,
NBER Working Paper No. 24288, Issued in February 2018

"Patent-Crawler, A real-time recursive focused web crawler to gather information on patent usage"
E. Orliac, G. Fourestey, SCITAS, EPFL
EPFL, 21st Dec.2017

"IPRoduct: Linking products to patents"
by Gaétan de Rassenfosse (EPFL),
Seminar at Scuola Superiore Sant’Anna , 14 November 2017, Pisa, Italy

"IPRoduct: Linking products to patents"
by Gaétan de Rassenfosse (EPFL),
AESIS Network webinar on “Measuring the Innovation Output of Science – Big data patterns & Signals for Science”, 24 October 2017, worldwide audience

"IPRoduct: Linking products to patents"
by Gaétan de Rassenfosse and David Portabella (EPFL),
2017 Summer School on Data & Algorithms for STI studies, 20–22 September 2017, Leuven, Belgium
Slides de Rassenfosse, Slides Portabella

"Big Data on HPC Clusters Tracing innovations in the marketplace"
by Vittoria Rezzonico, Gilles Fourestey (SCITAS, EPFL), Gaétan de Rassenfosse, David Portabella (IIPP, EPFL),
Fidis Unveiling, June 15th 2017

“Tracing Marketplace Innovations Using Big Data on HPC Clusters”,
Conference: High-Performance Computing (HPC) Advisory Council Switzerland Conference 2017,
Date: April 10-12, 2017
Place: Lugano, Switzerland

“IPRoduct: A database of linked patent-product pairs” by Gaétan de Rassenfosse,
Workshop: Symposium on innovation measurement and policy at the Japan Ministry of Education, Culture, Sports, Science and Technology
Date: February 8, 2017
Place: Tokyo, Japan

“IPRoduct: A database of linked patent-product pairs” by Gaétan de Rassenfosse,
Conference: Third Decennial OECD Blue Sky Forum on Science and Innovation Indicators
Date: September 19-21, 2016
Place: Ghent, Belgium

Software and data

Key parts of the IPRoduct project are publicly released under the license GNU General Public License v3.0.

Contact Us

We welcome any comments on the article and the data.

[email protected]