Disambiguation of Patent Inventors and Assignees Using High-Resolution Geolocation Data

Abstract

Patent data represent a significant source of information on innovation, knowledge production, and the evolution of technology through networks of citations, co-invention and co-assignment. A major obstacle to extracting useful information from this data is the problem of name disambiguation: linking alternate spellings of individuals or institutions to a single identifier to uniquely determine the parties involved in knowledge production and diffusion. In this paper, we describe a new algorithm that uses high-resolution geolocation to disambiguate both inventors and assignees on about 8.5 million patents found in the European Patent Office (EPO), under the Patent Cooperation Treaty (PCT), and in the US Patent and Trademark Office (USPTO). We show this disambiguation is consistent with a number of ground-truth benchmarks of both assignees and inventors, significantly outperforming the use of undisambiguated names to identify unique entities. A significant benefit of this work is the high quality assignee disambiguation with coverage across the world coupled with an inventor disambiguation (that is competitive with other state of the art approaches) in multiple patent offices.

The disciplinary effect of post-grant review – Causal evidence from European patent opposition

“…These findings are again identical when using the disambiguation byMorrison et al (2017). Our findings are also similar for European and non-European inventors.…”

“…In the appendix, we show that our main productivity results are robust to excluding outliers (such as the top 5% of inventors with respect to prior filings and technology areas) and to restricting the sample to inventors who patent both before and after the opposition outcome (Table C-1). In Table D-1, we also show that our results are unaffected by using the alternative inventor disambiguation by Morrison et al (2017). Finally, in Tables D-2 and D-3 we show that findings are very similar for European and non-European inventors.…”

“…17 The opposition division's decision can in principle be appealed on the ground of suspected lack of impartiality among the division members. However, there are only very few cases where this has occurred (see Gaessler et al 2017). correcting obvious errors using string similarity metrics, and (b) using the disambiguation provided by Morrison et al (2017) for robustness. For our primary dataset, we obtain information on 65,415 inventors associated with 29,009 first oppositions filed between 1994 and 2010.…”

“…The underidentification and weak identification tests are the heteroskedasticity-robust Kleibergen and Paap (2006) rk LM and Wald F statistics, respectively, as reported by the xtivreg2 Stata command (Schaffer, 2010). For an analogous table using the Morrison et al (2017) inventor disambiguation, see Table D-1 in the appendix. For analogous tables on the subsamples of European and foreign inventors, see Tables D-2 and D-3.…”

TM‐Link: An Internationally Linked Trademark Database

“…The Petrie and Julius (2019) classification algorithm obtained highly accurate results (F1 Score: 99.09 per cent, precision: 99.41 per cent, recall: 98.76 per cent), outperforming the previous world standard (Li et al 2014; Ventura et al 2015; Kim et al 2016; Morrison et al 2017; Yang et al 2017). …”