Analysis of Patent Data Sources in Adriatic Region

This presentation evaluates EPO PATSTAT as a data source for Adriatic countries, discussing the advantages and challenges compared to national patent offices. It explores the reliability, coverage, and integration possibilities of different data sources like WIPO and national patent offices, highlighting the importance of choosing the right source for patent data analysis.

• Patent Data
• EPO PATSTAT
• Adriatic Region
• Data Sources
• WIPO

waylen Follow

Uploaded on Aug 24, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

Presentation Transcript

1. iCRIOS Invernizzi Center for Research on Innovation, Organization and Strategy Patenting in Adriatic region data and coverage in PATSTAT Trieste 9/5/2016 Gianluca Tarasconi, iCrios DBA rawpatentdata.blogspot.com 1

2. In brief The goal of this presentation is to analyze EPO PATSTAT as data source for Adriatic area countries and its pro and cons. It will be also shown what source within patstat could be more reliable if national or international patent office 2

3. Why PATSTAT Obviously, where available the best data sources are national patent offices but some issues arise when trying to use the files to get one integrated DB (different formats, information missing, data unavailable for some authorities) and most of all the BI effort is very heavy. EPO Patstat offers an integrated source of data, a lot of plug and play extensions (ie nutsificatons, standard names ); most of all data can be compared to other patent offices or get some other data (ie triadics/PCT/characteristic of cited / citing) WIPO IP Statistics Data Center (on-line service enabling access to WIPO's statistical data) ,and WIPO Statistical Country Profiles for around 190 countries have a wider coverage but dont allow to build our own indicators. On the other hand data available have different reliability and coverage: limitations apply 5

4. Data from national patent offices (4 examples): Croatia: http://www.dziv.hr/en/e- services/on-line-database- search/patents/ good search engine no raw data Macedonia: http://www.ippo.gov.mk/Search/FederatedPat entRegister.aspx Limitation to ep/pct ; modest search nor raw data 4

5. Data from national patent offices (4 examples): Italy: http://brevettidb.uibm.gov.it/ good text search engine no raw data no search by inv / appl Serbia: http://89.216.38.53/rs- pubserver/search.jsp?lg=en modest search - no raw data WIPO pages links: http://www.wipo.int/ipstats/en/resources/ office_stats_reports.html 5

6. Data from national patent offices to Patstat: National offices convey to EPO local applications Timeframe, modalities and information shared can differ from office to office 6

7. Coverage check: a sheer publication count (I) Select a.APPLN_AUTH, a.FILING_YEAR, Count(a.APPLN_ID) From applications a Group By a.APPLN_AUTH, a.FILING_YEAR 7

8. Coverage check: a sheer application count (II) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 1 1 1 AL 4 8 4 7 2 7 7 11 8 7 7 5 8 6 5 BA 3363 2479 477 480 460 514 624 652 713 620 587 548 460 515 88 1 GR 720 782 867 882 1105 882 328 449 576 607 663 933 1001 1162 1116 534 HR 13069 13037 13095 12914 12353 12233 14582 13614 12397 12364 12479 12155 12283 11782 239 7 IT 55 67 79 105 87 18 31 64 6 2 1 4 ME 3 3 6 2 3 1 2 4 1 2 2 2 3 MK 19 28 48 67 77 466 581 705 645 497 543 334 224 152 61 RS 1418 1382 1346 2710 2527 2257 2178 1946 1736 1526 1328 922 565 436 85 5 SI Greatly differ from official data (caveat: data transmission lag to EPO can differ from office to office so comparisons amons very recent data can lead to misguiding results) HR data source: http://www.dziv.hr/files/File/go-izvjesca/godisnje_izvjesce_2013.pdf 8

9. Percentage of granted patents and PCT (II) Granted patents = application has passed examination and succesfully reached grant phase; PCT = applied at local office, for PCT (Patent Cooperation Treaty - wipo) route (see http://www.wipo.int/pct/en/) (SQL CODE IN APPENDIX) 9

10. Percentage of granted patents and PCT (II) granted % patoff AL BA GR HR IT ME MK RS SI 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 0 0 0 0,0% 95,2% 36,3% 70,1% 0,0% 0,0% 0,0% 22,1% 0,0% 91,7% 38,1% 0,2% 0,0% 0,0% 0,0% 23,0% 0,0% 67,1% 39,7% 0,1% 0,0% 0,0% 0,0% 21,0% 0,0% 69,0% 37,8% 0,1% 0,0% 0,0% 0,0% 9,3% 0,0% 69,4% 31,2% 0,4% 0,0% 0,0% 0,0% 10,4% 0,0% 75,7% 30,6% 0,1% 0,0% 0,0% 0,0% 11,7% 0,0% 68,0% 41,2% 0,1% 0,0% 0,0% 0,0% 10,6% 0,0% 67,8% 56,6% 0,1% 0,0% 0,0% 0,0% 13,7% 0,0% 66,2% 69,1% 0,1% 0,0% 0,0% 0,0% 14,7% 0,0% 62,3% 28,7% 0,0% 0,0% 0,0% 0,0% 18,7% 0,0% 55,0% 15,8% 0,0% 0,0% 0,0% 0,0% 25,8% 0,0% 51,6% 9,8% 0,0% 0,0% 0,0% 55,9% 7,7% 0,0% 0,0% 52,6% 1,7% 0,0% 0,0% 53,4% 1,1% 0,0% 0,0% 0,4% 0,0% 0,0% 0,0% 53,8% 0,0% 0,0% 48,2% 100,0% 0,0% 39,4% 0,0% 74,1% pct % patoff AL BA GR HR IT ME MK RS SI 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 100,0% 2011 2012 100,0% 100,0% 2013 2014 2015 100,0% 1,2% 5,3% 4,0% 0,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 2,1% 2,4% 1,9% 1,7% 1,9% 1,6% 87,5% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 100,0% 1,7% 13,2% 12,1% 11,1% 6,4% 9,9% 9,4% 5,9% 7,0% 7,4% 5,4% 6,8% 10,4% 12,5% 4,9% 6,1% 6,3% 5,7% 6,2% 5,9% 6,6% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 9,0% 5,9% 6,5% 0,0% 9,2% 4,9% 4,6% 0,0% 7,5% 5,0% 4,1% 0,0% 8,8% 4,4% 3,4% 0,0% 8,9% 2,0% 3,2% 11,5% 3,1% 3,0% 45,5% 100,0% 1,5% 99,6% 100,0% 0,0% 100,0% 5,8% 11,3% 100,0% 8,2% 51,8% 10 0,0% 1,7% 2,8% 2,2% 5,1% 3,9% 3,8% 4,3% 2,0% 5,2% 3,3% 7,9% 11,8% 19,5% 0,0%

11. % of granted and PCT patents (III) Grants data: some countries do not update data after application (IT, RS,BA, MK ) some have unrealistic % (SI) PCT %: 100% means only PCT data in patstat; note last 2 years data: means transmission lag between epo and national offices 11

12. What then? National data not reliable Usage of PCT data + reliable, homogenuous - bias toward high expected value - address quality bad ( enrichment with OECD regpat http://www.oecd.org/sti/inno/oecdpatentdatabases.htm 12

13. Using PCT data cy AL BA GR HR IT ME MK RS SI 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 8 8 1 1 7 1 2 4 6 9 5 7 9 15 107 74 3560 13 147 70 3623 10 130 56 3441 9 12 111 51 3586 7 74 56 92 57 108 85 2398 105 85 2633 78 81 128 73 3208 140 52 3397 120 60 3434 136 48 3544 80 27 2041 2250 2856 1966 4 1 1 3 4 3 3 8 3 7 5 3 3 2 4 3 2 17 113 30 129 30 151 27 140 25 154 17 139 32 130 25 81 81 60 92 82 92 121 Counting PCT patents by inventors country 13

14. TOP 20 patenting regions By NUTS2 Italy excluded priority year > 2000 Ctry_code SI SI SI SI SI SI SI GR GR SI SI GR GR GR GR SI SI GR SI region ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ZAHODNA SLOVENIJA ATTIKI ATTIKI ZAHODNA SLOVENIJA VZHODNA SLOVENIJA ATTIKI ATTIKI ATTIKI ATTIKI VZHODNA SLOVENIJA ZAHODNA SLOVENIJA ATTIKI VZHODNA SLOVENIJA earliest_filing_year napp 2008 2011 2009 2010 2012 2007 2006 2007 2012 2005 2010 2005 2008 2009 2010 2007 2002 2011 2005 126 122 114 113 105 101 98 96 92 89 88 85 80 80 78 78 77 71 71 14

15. Italian Adriatic regions at their best Earliest filing year Ctry_code region EMILIA- ROMAGNA napp IT 2007 626 IT VENETO FRIULI-VENEZIA GIULIA 2009 515 IT 2012 220 IT MARCHE 2006 127 IT PUGLIA 2007 87 IT ABRUZZO 2008 69 IT MOLISE 2009 8 Best y2k: LOMBARDIA 2006 - 1211 15

16. Copatenting Copatenting between 2 countries with 3 or more patents in one year Ctry_code1 Ctry_code2 GR HR IT IT IT IT IT IT IT IT IT IT Count EARLIEST_FILING_YEAR IT SI GR GR GR GR HR HR HR RS SI SI 7 5 8 4 4 4 3 3 3 3 3 3 2011 2006 2012 2005 2009 2013 2001 2009 2010 2009 2007 2009 16

17. Most patented technologies By ipc4, country, after y2k (sql code in appendix) Count appln id cpc4 Ctry_code Description A61K IT 4489PREPARATIONS FOR MEDICAL, DENTAL, OR TOILET PURPOSES A61K SI 310PREPARATIONS FOR MEDICAL, DENTAL, OR TOILET PURPOSES A61K GR 228PREPARATIONS FOR MEDICAL, DENTAL, OR TOILET PURPOSES A61K HR 183PREPARATIONS FOR MEDICAL, DENTAL, OR TOILET PURPOSES DIGITAL COMPUTERS IN WHICH AT LEAST PART OF THE COMPUTATION IS EFFECTED ELECTRICALLY; ARRANGEMENTS FOR HANDLING DIGITAL DATA G06F RS 22 F03D BA 8WIND MOTORS PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, E.G. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS A61F MK 4 A61C ME 3DENTISTRY; APPARATUS OR METHODS FOR ORAL OR DENTAL HYGIENE SKATES; SKIS; ROLLER SKATES; DESIGN OR LAYOUT OF COURTS, RINKS OR THE LIKE A63C AL 3 17

18. Knowledge spillovers Citations between countries citing cited Count of citation SI IT 75 GR IT 70 IT SI 51 IT GR 45 HR IT 43 IT HR 23 GR SI 15 SI HR 8 HR SI 7 RS IT 6 BA IT 3 GR HR 3 18

19. Conclusions National data are not reliable in patstat PCT data have a bias towards tech and applicants with international propension Comparison can be done by groups (small vs small / big vs big) 19

20. Appendix 1: SQL code for PATSTAT (I) create table applications Select APPLN_ID, APPLN_AUTH, APPLN_NR, APPLN_KIND, year(appln_filing_date) As FILING_YEAR, APPLN_AUTH As REFAPPLN_AUTH, "N" As GRANTED, IF(appln_kind = "W", "Y", "N") as PCT From patstat.tls201_appln T01 where appln_auth in ('AL','BA','GR','HR','IT','ME','MK','RS','SI ) and appln_kind <> "D2" and year(appln_filing_date) <>9999; -- Removes unpublished delete a.* From applications a Left Join patstat.tls211_pat_publn b On a.APPLN_ID = b.APPLN_ID where b.appln_id is null; 20

21. Appendix 1: SQL code for PATSTAT (II) -- sets grant flag update applications a Inner Join patstat.tls211_pat_publn b On a.APPLN_ID = b.APPLN_ID set a.GRANTED = "Y" Where b.PUBLN_FIRST_GRANT = 1; -- PCT / TRIADIC / GRANTED CALC Select a.FILING_YEAR, a.REFAPPLN_AUTH, Sum(If(a.GRANTED = 'Y', 1, 0)) / Count(a.APPLN_ID) As gr, Sum(If(a.PCT = 'Y', 1, 0)) / Count(a.APPLN_ID) As pct From applications a Group By a.FILING_YEAR, a.REFAPPLN_AUTH 21

22. Appendix 1: SQL code for PATSTAT (III) 22

23. Appendix 1: SQL code for PATSTAT (IV) -- ipc4 by country year use test2; drop table if exists t1; drop table if exists t6; create table t6 as select * from otherdbs.pct_inv_reg where CTRY_CODE in ('AL','BA','GR','HR','IT','ME','MK','RS','SI'); create table t1 as select * from patstat.tls201_appln where APPLN_KIND = 'W'; alter table t6 add index i1(appln_id); alter table t1 add index i1(appln_id); --count by cy ipc4 create table cpc4_by_cy_year Select Left(t24.ipc_class_symbol, 4) As ipc4, test2.t6.Ctry_code, Count(Distinct t24.appln_id) As Count_appln_id From test2.t1 Inner Join patstat.tls209_appln_ipc t24 On t24.appln_id = test2.t1.APPLN_ID Inner Join test2.t6 On test2.t1.APPLN_ID = test2.t6.Appln_id Where test2.t1.EARLIEST_FILING_YEAR > 1999 Group By Left(t24.ipc_class_symbol, 4), test2.t6.Ctry_code, test2.t1.EARLIEST_FILING_YEAR; 23