patentics com patent ntics ics syst stem
play

Patentics.com Patent ntics ics Syst stem First largest Concept - PowerPoint PPT Presentation

Patentics.com Patent ntics ics Syst stem First largest Concept Search Engine in the World; Mathematically Models 6.2M US, 3M EP/WO, 2.9M CN Full-Text Patent Documents; Performance Matches with Worlds Patent Examiners! Intelligent Auto


  1. Patentics.com

  2. Patent ntics ics Syst stem First largest Concept Search Engine in the World; Mathematically Models 6.2M US, 3M EP/WO, 2.9M CN Full-Text Patent Documents; Performance Matches with World’s Patent Examiners’! Intelligent Auto Search;  Intelligent Data Contents with 8 billions of context-aware hyperlinks;  For World or World P Pate tent Auth uthor orities: 1. 1. 20% 35% improvement in e -xamination process; No Office Space required! No Headcount increased! For Da or Data ta S Serv rvice ce Prov rovid iders: 2. 2. Connected/Structualized Contents Space; 300% Users Click through Rate/Pageview; Service Revenue Increased!

  3. Input patent number and get ranking results Ranking first matched with examiner cited X Ranking second matched with examiner cited X Ranking eighth matched with examiner cited X

  4. EPO Search Report cited 3 US documents as X and all matched in first screen by Patentics

  5. Autom omatic M tic Matc tches with with Se Search ch Reports b ts by World ld’s P s Pate tent Examiner ners  Input a patent number without any search strategies, Calculate and Rank the related patent documents automatically;  Match cited X/Y/A in search reports submitted by EPO, USPTO, JPO, SIPO examiners through PCT;  English Search Report Contest-1: EPO2007-08 total 73,282 with cited US/US family X documents 34,213; 1. USPTO2007-08 total 44,847 with cited US/US family X documents 22,480; 2. JPO2004-09 total 89,683 with cited US/US family X documents 8,623; 3. JPO2004-09 total 89,683 with Abstract and cited US/US family X documents 3,475; 4.  Chinese Search Report Contest-1: SIPO2004-09 total 15,522 with cited CN/CN family X documents 6,791; 1.

  6. Patentics Top 20 (first screen) X Document Hit Ratio Distribution 4.00% 3.00% Ranking Hit Ratio f(i) 2.00% 1.00% 0.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 US2007-08 Ranking Position (i) Just input a patent number, let Patentics calculates and ranks, 1. Most probable position ranked by Patentics for cited X documents is at top 1; 2. Higher than 3% chance, cited X documents is at top 1.

  7. Cumulative Distribution of Patentics Top 20 (first screen result) matches with Cited X Documents 15.00% Cumulative Distribution F(p) 10.00% 5.00% 0.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 US2007-08 Ranking Interval (p) Just input a patent number, let Patentics calculates and ranks, 1.Higher than 15% chance, cited X documents will appear in first result screen as ranked by Patentics .

  8. Cumulative Distribution of Patentics Top 400 matches with Cited X, Y, A Document 45.00% 40.00% ion F(p) 35.00% Distributio 30.00% ve Distribu 25.00% 20.00% Cumulative 15.00% Cumulat 10.00% 5.00% 0.00% 0 50 100 150 200 250 300 350 400 Interval (p) US2007-08 X Document Y Document A Document 1. After reading through documents, examiners cite X/Y/A documents and make a grading on relevance in the order of X, Y/A; 2. The same relevance grading is acquired and emulated by Patentics, X is highest, Y/A is low.

  9. Trilateral Patentics Top 400 Cit ited X Do Docume uments C Compa mparison 45.00% 40.00% 35.00% (p) ion F(p 30.00% ibutio 25.00% Distrib ive D 20.00% lativ Cumula 15.00% 10.00% 5.00% 0.00% 0 50 100 150 200 250 300 350 400 Rank nking position ( n (p) EPO2007-08/732821/34213 USPTO2007-08/44847/22480 JPO2005-08/60229/7469 * ORG/from year-to/Total Count/Cited Count

  10. Patenti tics cs To Top 20 20 (fi (first st sc screen een) X Docu cument Hit Hit Ra Ratio io Dis istrib ribution 5.00% 4.00% f(i) Hit Ratio 3.00% Ranking H 2.00% 1.00% 0.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 SIPO2004-09 Ranking Position (i) Just input a patent number, let Patentics calculates and ranks, 1. Most probable position ranked by Patentics for cited CN X documents is at top 1; 2. Near 5% chance, cited CN X documents is at top 1.

  11. Cumula lativ tive Distribu tributi tion on o of Patentic tics To Top 20 20 (first s scr creen een res esult) ma matc tches w with ith Cite ited X d X Doc Docume ments 22.00% 20.00% 18.00% Cumulative Distribution F(p) 16.00% 14.00% 12.00% 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 SIPO2004-09 Ranking Interval (p) Just input a patent number, let Patentics calculates and ranks, 1.Higher than 20% chance, cited CN X documents will appear in first result screen as ranked by Patentics .

  12. Cumulative Distribution of Patentics Top 400 matches with Cited X, Y, A Document 55.00% 50.00% 45.00% ion F(p) 40.00% Distributio ve Distribu 35.00% 30.00% 25.00% Cumulative 20.00% Cumulat 15.00% 10.00% 5.00% 0.00% 0 50 100 150 200 250 300 350 400 Interval (p) SIPO2004-09 X Document Y Document A Document After reading through documents, examiners cite CN X/Y/A documents and make a grading on relevance in the order of X, Y/A; The same relevance grading is acquired and emulated by Patentics, X is highest, Y/A is low.

  13. Patentic Pa ics Intel telligence Al Align gns Best est with th th the e Best est Huma man Intel telligence A hypothesis in academic that fundamental innovation patents and litigated patents are good quality patents comparing to general ones, because 1. Inventors make fundamental innovation researching prior art with great effort either citing extensively or nothing; 2. Inventors anticipate litigation trying to make patents as strong as possible by researching and citing prior art extensively; Challenge is, can Patentics validate that well-known hypothesis with mathematical precision 1. Based on a list of 667 fundamental US patents provided by IEEE as a test sample set (Fundamental Patents), and randomly sample 667 patents with the same IPC and filed in the same month as a comparison set (Comparison Patents-F); 2. Based on a list of 300 litigated US patents provided by http://www.patstats.org as a test sample set (Litigated Patents), and randomly sample 300 patents with the same IPC and filed in the same month as a comparison set (Comparison Patents-L); The re result is is, , Paten tentics intel telligence ma matc tches not ot on only with th th the e huma man intel elligen ence, e, but t also so best est with th th the e best est huma man intel telligence. References : 1. Valuable Patents, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=426020&rec=1&srcabs=881842 2. Characteristics of patent litigation: a window on competition, http://www.rje.org/abstracts/abstracts/NBER/rje_Spring'01_Lanjouw.pdf 3. The Influence of IEEE on Key Patents, http://www.ieee.org/portal/cms_docs_iportals/discover/sub_pages/IEEE_Key_Patents_2006.pdf

  14. Cumulativ ive D Dis istrib ributio ion o of f Patent entics cs Top 4 p 400 matc tche hes wi with pr h prior or art c cite ted by d by Funda undament ntal Patent ents/Comparison P n Patents-F’s i s inve ventors rs 100.00% 90.00% 80.00% F(p) p) ion F( 70.00% ributio 60.00% Distribu 50.00% lative Dis 40.00% Cumula 30.00% 20.00% 10.00% 0.00% 0 50 100 150 200 250 300 350 400 Interval al (p) Fundamental Patents/667 Comparison Patents-F/667 Patentics Intelligence Aligns Best with the Best Human Intelligence

  15. Cumulativ ive D Dis istrib ributio ion o of f Patent entics cs Top 4 p 400 matc tche hes wi with pr h prior or art c cite ted by d by liti tigated Patent ents/Comparison P n Pate tents-L’ L’s i s inve ventors rs 100.00% 90.00% 80.00% F(p) ion F( 70.00% ributio 60.00% Distribu 50.00% lative Dis 40.00% Cumula 30.00% 20.00% 10.00% 0.00% 0 50 100 150 200 250 300 350 400 Interv rval (p) p) Litigated Patents/300 Comparison Patents-L/300 Patentics Intelligence Aligns Best with the Best Human Intelligence

  16. Cro ross ss-Language S e Sea earch ching From our study, if a foreign patent is cited, higher than 70%, it is a 1. US patent/application; Leverage highly accurate 6.2 Million US Patent Canonical Model 2. (UPCM); By Inputting a non-US patent number (CN Application) directly, and 3. searching in UPCM; Comparison Tests conducted, 4. Utilizing CN application’s US family number to search in UPCM and match ranking result with EPO examiners ◦ search report (EN->EN); EN Directly inputting CN application number to search in UPCM and match ranking result with EPO examiners ◦ search report (CN->EN); CN

  17. Cumula lativ ive D Distrib ibutio ion o of Cumula lativ ive D Distrib ibutio ion o of CN CN->EN T Top 4 400 Matches EN EN->EN EN Top 400 M Matches 25.00% 40.00% 35.00% 20.00% 30.00% p) F(p) F( ion p) F(p) 25.00% ributio 15.00% F( Distribu ion ributio 20.00% lative Dis Distribu 10.00% lative Dis Cumula 15.00% Cumula 10.00% 5.00% 5.00% 0.00% 0.00% 0 100 200 300 400 0 100 200 300 400 Interv rval (p) p) Interval al (p) EN->EN CN->EN SIPO2004-09/15522/1049

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend