Patentee Name Harmonisation - ecoom.be
Patentee Name Harmonisation - ecoom.be
Patentee Name Harmonisation - ecoom.be
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
DIAGNOSTICS<br />
MITSUBISHIJUKOGYO MITSUBISHIKASEI 6 0,35 ≈<br />
ALLERGANSALES ALLERGAN 5 0,38 ≈<br />
CATERPILLAR CATERPILLARTRACTOR 7 0,39 ≈<br />
KONINKLIJKEPHILIPSELECTRONIC PHILIPSELECTRONIC 11 0,39 =<br />
CENTRENATIONALDELARECHERCHE<br />
SCIENTIFIQUE<br />
ETABLISSEMENTPUBLICDITCENTRE<br />
NATIONALDELARECHERCHE<br />
SCIENTIFIQUECNRS<br />
26 0,40 =<br />
If correct matches were to coincide with distinctive values in terms of the distance<br />
measures, automated procedures could <strong>be</strong> envisaged. However, the results of this exercise do<br />
not suggest such a pattern. An inspection of the distance values and validation reveals that<br />
defining an unambiguous criterion value is not so straightforward. Moreover, one observes<br />
immediately that several cases require a more in-depth analysis in order to define whether or<br />
not both patentees are similar. Such an assessment immediately extends <strong>be</strong>yond name<br />
harmonizing per se and is <strong>be</strong>st categorized as an exercise in legal entity harmonization (see<br />
section 2 - <strong>Patentee</strong> name harmonization and legal entity harmonization). As this approach is<br />
clearly <strong>be</strong>yond the scope of the objectives envisaged in this contribution, it has not <strong>be</strong>en<br />
included in the final methodology. At the same time, it goes without saying that enriching<br />
approximate string searching with address information seems highly promising when striving for<br />
legal entity harmonization and might also contribute to automated name harmonizing. Both<br />
extensions do imply however considerable validation efforts.<br />
7 CONCLUSION<br />
In this contribution, we have developed a comprehensive approach oriented towards<br />
name harmonizing. Emphasis has <strong>be</strong>en placed on maximizing the accuracy of procedures that<br />
can <strong>be</strong> implemented automatically, i.e. without additional - and time-consuming - validation<br />
efforts that require secondary information sources. <strong>Name</strong> variations are not combined if there is<br />
any doubt that the names relate to different legal entities.<br />
This has resulted in a transparent method whose outcome has <strong>be</strong>en a reduced set of<br />
harmonized names. At this stage, the total num<strong>be</strong>r of patentee names has <strong>be</strong>en decreased by<br />
17%. 13.5% of the harmonized names are matched with more than one original patentee<br />
name, matching 2.6 names on average. When harmonizing takes place, the num<strong>be</strong>r of patents<br />
allocated to the same entity increases on average by 7.2 patents in absolute terms (33.1% in<br />
relative terms), signaling a considerable impact.<br />
A detailed validation exercise was conducted for 35 harmonized names. Findings<br />
revealed accuracy levels of 100% percent and a level of completeness of 99,62%. More details<br />
about this validation can <strong>be</strong> found in Appendix 7.<br />
EUROSTAT and its partners deli<strong>be</strong>rately opted for a transparent method so that all<br />
interested parties will <strong>be</strong> able to build further on the results obtained. In the <strong>be</strong>lief that the<br />
procedures descri<strong>be</strong>d in this methodology can <strong>be</strong> further enriched and refined – and this also<br />
applies to legal entity normalization - we would encourage activities in this direction.<br />
Improving the accuracy levels for the methodology as a whole is feasible by introducing<br />
expert assessments in a systematic manner. Given the volume of names involved, such an<br />
effort is <strong>be</strong>yond the current resources of EUROSTAT and its partners who developed this<br />
methodology (INCENTIM/SOOS, K.U.Leuven, and SOGETI). At the same time, numerous<br />
researchers and analysts are currently working on name harmonizing efforts with specific<br />
samples (e.g. technological fields, countries/regions, and sectors). For researchers engaged in<br />
such efforts, building on this methodology might <strong>be</strong> helpful; equally, the insights obtained by<br />
researchers and analysts might <strong>be</strong> <strong>be</strong>neficial for further refinement of the current methodology.<br />
In other words, by sharing the methodology developed among the different communities<br />
involved in patentee analysis, further improvements could <strong>be</strong> envisaged. Consequently,<br />
EUROSTAT and its partners decided to put the complete methodology into the public domain.<br />
In Appendix 1 and Appendix 2, the full set of procedures is made available, making<br />
allowances for implementation, verification, and the development of appropriate<br />
extensions. Furthermore, given its continuous involvement in the PATSTAT Taskforce activities,<br />
16