Recognizing Inventors with a Little Help from CRIOS

SINCE 2001 A TEAM OF THE RESEARCH CENTER HAVE BEEN CLEANING AND STANDARDIZING RAW DATA PROVIDED BY THE EUROPEAN PATENT OFFICE ON 120 MLN PATENTS. THANKS TO THEIR WORK WE CAN UNDERSTAND WHO PATENTS WHAT IN ITALIAN UNIVERSITIES

If the Agenzia nazionale di valutazione del sistema universitario e della ricerca (ANVUR), the Italian agency for research assessment, can use reliable data about patents in its effort to assess universities, part of the merit is to be ascribed to Bocconi’s Center for Research in Innovation, Organization and Strategy (CRIOS) that, in January 2016, will provide ANVUR with a second, more complete, batch of data, after that provided in 2015.
 
An internationally renowned team of CRIOS researchers have been working on patent data collection and analysis for a long time. Since 2001 Francesco Lissoni, Stefano Breschi and database manager Gianluca Tarasconi have been cleaning and standardizing raw information on inventors provided by the European Patent Office (EPO) so that it can be used in large-scale statistical exercises.
 
“In its PATSTAT database, EPO collects data in a way that can be useful to the examiners, but there isn’t any harmonization rule”, Tarasconi says. “If I were an inventor, my name could be written Gianluca, Gian Luca or G. Luca, for instance, and an automatic research would assign my inventions to three different persons. We have, thus, cleaned, and are continuously updating, data about all the patents registered after 1978, something like 120 mln patents from 90 patent offices”. It’s a titanic work, impossible to realize without the help of text mining and data mining tools.
 
Now it’s possible to cross the cleaned data with data from other databases and discover, for example, the average size of patenting companies or whether patent registration tends to come with scholarly publication. The PATSTAT-CRIOS database is consequently used by many patent researchers. “We’ve been data providers”, Tarasconi continues, “for MIT and Harvard researchers and for European projects aiming to understand the workings of innovation in the Continent”.
 
“Our data will be used by ANVUR”, Lissoni explains, “in the second edition of their four-year assessment of Italian research (Vqr 2011-2014). Now they want to assess the so-called third mission, that is direct knowledge transfer not using the medium of research and teaching. Among these activities, technology transfer via markets is paramount and patents are a relevant but not easy to elaborate indicator. There are difficulties in collecting data and in assigning patents to inventors and institutions”.
 
Until 2014 ANVUR used to rely on inventors’ notifications. In 2015 CRIOS provided standardized data on universities’ inventors, in January 2016 data will encompass also research centers.

by Fabio Todesco