Nal 2011 05-19

674 views
621 views

Published on

Presentation at NAL

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
674
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
9
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • Note: we identified outlinks to RAMAEU and GEMET, and they have taken them as inlinks to their own thesaurus.
  • - All links are checked by a domain expert.
  • - All links are checked by a domain expert.
  • Ifresources are marked up withsemanticallydefined and machinereadableconcepts, they can belinked and mashed up preciselyaswehaveseen in the examplefrom the BBC.In thisexamplewe start withan AGRIS record on Hazardouswaste, whichisindexedwith AGROVOC. Alreadynowwe can easily link to material indexedwithEurovoc, hereanexamplefromEuroLex. If the UNBIS thesaurus wouldberestructuredto a conceptscheme and publishedas LOD, related UN documentscouldbeattachedautomaticallyby the machine.
  • Ifresources are marked up withsemanticallydefined and machinereadableconcepts, they can belinked and mashed up preciselyaswehaveseen in the examplefrom the BBC.In thisexamplewe start withan AGRIS record on Hazardouswaste, whichisindexedwith AGROVOC. Alreadynowwe can easily link to material indexedwithEurovoc, hereanexamplefromEuroLex. If the UNBIS thesaurus wouldberestructuredto a conceptscheme and publishedas LOD, related UN documentscouldbeattachedautomaticallyby the machine.
  • Ifresources are marked up withsemanticallydefined and machinereadableconcepts, they can belinked and mashed up preciselyaswehaveseen in the examplefrom the BBC.In thisexamplewe start withan AGRIS record on Hazardouswaste, whichisindexedwith AGROVOC. Alreadynowwe can easily link to material indexedwithEurovoc, hereanexamplefromEuroLex. If the UNBIS thesaurus wouldberestructuredto a conceptscheme and publishedas LOD, related UN documentscouldbeattachedautomaticallyby the machine.
  • Ifresources are marked up withsemanticallydefined and machinereadableconcepts, they can belinked and mashed up preciselyaswehaveseen in the examplefrom the BBC.In thisexamplewe start withan AGRIS record on Hazardouswaste, whichisindexedwith AGROVOC. Alreadynowwe can easily link to material indexedwithEurovoc, hereanexamplefromEuroLex. If the UNBIS thesaurus wouldberestructuredto a conceptscheme and publishedas LOD, related UN documentscouldbeattachedautomaticallyby the machine.
  • How does this work: A resource is connected with each concept URI in the web. The concepts between three vocabularies are having same literal which is connected with owl:sameAS/exactMatch relationship. As we are speakingaboutthesauri and notontologieswekept the relation tobechosenpurposelyvague. The conceptscouldbematchedwithowl:sameAS or the termscouldbematcheswith SKOS:exactMatch. A lotofdiscussion on thisisongoing
  • Oneof the groundbreakingenterprises in this area isThomsonReuters “Open Calais”. Thisis a webservicethatprovidessemanticmark up foranyunstructured text thatyoufeedintotheir service The service is free ofCharge. Why? I will show youlater.
  • My team in collaborationwith the IndianInstituteofTechnology in Kanpur isdeveloping a similar service foroursubject area.
  • Wehavehere a text from 1964 without a bibliographic record at handabout a plantprotectionissue
  • Open Calais isverygood in thoseareas, in whichtheyhavetheirownelaboratedconceptschemeagainstwhich the texts are analyzed: “Places”, “Persons”, “Business Processes” , “IndustryTerms”, butitisweak in the specifictopicanalysis, whattheycall “social tags”
  • AgroTaggerstilllacksmanyof the sophisticated featuresof “Open Calais” ,butismuch, muchbetter in the subjectanalysisof the text
  • The chart represent the records published from the AGRIS partner institutions of the relevant Country (Scielo is the exception being a service provider itself)The citation column is the number of references of articles from scientific journals.
  • Nal 2011 05-19

    1. 1. AGROVOC Linked Open Data and the VocBench, potentials for the community<br />Dr. Johannes Keizer<br />Office ofKnowledge Exchange, Research and Extension<br />Food andAgricultureOrganizationofthe UN<br />Talk atthe US National Agricultural Library,<br /> May 20 2011<br />
    2. 2. We will promote research for food and agriculture, including research to adapt to, and mitigate climate change, and access to research results and technologies at national, regional and international levels. <br />We will reinvigorate national research systems and will share information and best practices. <br />We will improve access to knowledge.<br />worldfoodsummit 2009<br />
    3. 3. …Sharing Information<br />1975 – AGRIS<br />1980 - AGROVOC<br />2000 - Rethinking of AGRIS<br />2002 - AOS/AGMES<br />2005 - CIARD<br />2005 - AIMS<br />
    4. 4. http://aims.fao.org<br />
    5. 5. Todays Talk <br />What’s new with AGROVOC<br />AGROVOC Linked Open Data<br />The VocBench<br />Why mapping Vocabularies<br />Automatic Indexing<br />AGRIS<br />
    6. 6. AGROVOC<br />
    7. 7. AGROVOC<br />A multilingual agricultural vocabulary organized as concept scheme in 20 languages<br />Covers agriculture, forestry, fisheries and related themes (food security, land use, environment, etc.)<br />Organized in sub-vocabularies, e.g. chemicals, fisheries terms, scientific/common names of organisms<br />Maintained by a global community (e.g. librarians, terminologists, information managers) using VocBench<br />
    8. 8. AGROVOC - Statistics <br />
    9. 9. AGROVOC - Restructuring <br />Goal: Transform AGROVOC from a traditional thesaurus into a concept scheme with distinction between conceptual level and terminological level<br />Overall revision done by FAO in collaboration with KSI (Knowledge Sharing and Innovation) team at ICRISAT, Hyderabad, India<br />Top concepts reduced from 918 to 25<br />Around 85,000 term relations revised<br />Non-hierarchical relationships refined by semantic relations<br />Ca. 4,000 non-preferred terms changed to preferred terms<br />
    10. 10. Top concepts<br />
    11. 11. Relationships (examples)<br />
    12. 12.
    13. 13. Thesauri into the AGROVOC LOD Cloud <br /><ul><li> 18000 outlinks
    14. 14. 2000 inlinks</li></ul>EUROVOC<br />NALT <br />AGROVOC<br />RAMEAU<br />GEMET <br />STW <br />LCSH<br />
    15. 15.
    16. 16.
    17. 17.
    18. 18. AGROVOC LOD-inlinks <br />Trusted Links from <br />AGROVOC <br />
    19. 19. AGROVOC Links after 3 weeks LOD<br />Outlinks:<br />GEMET-AGROVOC 1,198<br />RAMEAU-AGROVOC  :700<br />Total Outlinks: 1898<br />Inlinks:<br />AGROVOC-EUROVOC:1,297<br />AGROVOC-GEMET:1,198<br />AGROVOC-LCSH :1,093<br />AGROVOC-NAL: 13,390<br />AGROVOC-STW:1136<br />AGROVOC-RAMEAU:700<br />Total Inlinks:18,814<br />
    20. 20.
    21. 21. Get the NALT into the LOD cloud too!<br />Some notes from Ahsan<br />Issues<br />1.  Only  problem is that  the online version  and  the SKOS version 2011 of NALT are  totally different according to the terms code .  They need to make the stable version of their terms code .2. Since  the AGROVOC is connected with the  NALT so they are in the LOD according to the principle. They can put our 13,000 mapping links into their SKOS file. It will cover outlinks and inlinks.<br />Necessary activities<br />1. Put the SKOS version in the triple store.2. Make the dereferenceable URIs in their website (They can use the biotech drupal module that we are going to build now)3. Publish the URIs  by using Pubby  tool      http://www4.wiwiss.fu-berlin.de/pubby/<br />
    22. 22. Europe:(It is better to use this example during the presentation)http://aims.fao.org/aos/agrovoc/c_2724From the Top concept:Ref:  http://aims.fao.org/aos/agrovoc/c_7644Vocbench (Production)Ref:   http://agrovoc.mimos.my/vocbenchv1.1i/VocBench(Sandbox)Ref:http://agrovoc.mimos.my/vocbenchv1.1i/<br />
    23. 23. The VocBench<br />
    24. 24. The VocBench<br />VocBench<br />concepts and entitiestriples<br />
    25. 25. VocBench Features<br /><ul><li> Domain independent
    26. 26. Structure independent (i.e. thesauri, Glossaries, etc)
    27. 27. Supports RDF (SKOS, SKOS-XL), OWL
    28. 28. Supports collaborative editing
    29. 29. Supports editorial workflow, with user roles
    30. 30. Simple and advanced search
    31. 31. Supports data export: SKOS, Relational format (MySQL) </li></li></ul><li>
    32. 32. Why linking vocabularies?<br />
    33. 33.
    34. 34. http://aims.fao.org/aos/agrovoc/c_7825<br />
    35. 35. http://eurovoc.europa.eu/218754<br />http://aims.fao.org/aos/agrovoc/c_7825<br />
    36. 36. http://eurovoc.europa.eu/218754<br />http://aims.fao.org/aos/agrovoc/c_7825<br />
    37. 37. http://eurovoc.europa.eu/218754<br />http://agclass.nal.usda.gov/nalt/2011.xml#1780<br />http://aims.fao.org/aos/agrovoc/c_7825<br />
    38. 38. Linking data through common URIs<br />TOXIC SUBSTANCES<br />http://www.agnic.org/search/CAT85822953<br />UNBIS<br />AGROVOC<br />NALT<br />http://aims.fao.org/aos/agrovoc/c_7825<br />http://agclass.nal.usda.gov/nalt/2011.xml#1780<br />http://eurovoc.europa.eu/218754<br />Eurovoc<br />http://agris.fao.org/agris-search/search/display.do?f=1996/TR/TR96001.xml;TR9600026<br />http://unbisnet.un.org:8080/ipac20/ipac.jsp?session=128F308557F34.283092&profile=bib&uri=full=3100001~!685149~!1&ri=1&aspect=subtab124&menu=search&source=~!horizon<br />http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2010:202:0011:0015:EN:PDF<br />http://aims.fao.org/aos/agrovoc/c_12332 owl:sameAshttp://eurovoc.europa.eu/219871<br />skos: exact match UNBIS: Toxic Substances<br />
    39. 39. If all institutions, which publish about toxic wastes would:<br /><ul><li>- Index their publications with URIs from AGROVOC,GEMET, NALT, LCSH or EUROVOC
    40. 40. (many do – low hanging fruit!)
    41. 41. - Publish their metadata as LOD
    42. 42. (quite easy to do, bibData map well to RDF</li></ul>Then<br />Everyone who knows to write SparqlQeries could get all these publications with one shot for a new website on toxic wastes<br />
    43. 43. Vocabularies and LOD<br />Simply publishing your data as RDF does not link them to other data sets <br />Creating this links by humans is interesting in detail, but unrealistic as mass processing<br />Linking 2 standard vocabularies can link 200 datasets which use these standard vocabularies<br />
    44. 44. AgroTagger<br />And<br />OpenCalais<br />
    45. 45.
    46. 46. <ul><li>Does Concept identification in unstructured texts
    47. 47. Uses Agrovoc as a controlled vocabulary
    48. 48. Prototype under testing with excellent results (entire repository of ICARDA indexed)
    49. 49. Will produce in future Structured RDF files that can be used to link data like “open Calais”</li></ul>AgroTagger<br />
    50. 50.
    51. 51.
    52. 52.
    53. 53. AGRIS<br />
    54. 54. Oneyear<br />
    55. 55. AGRIS repositoryData aggregation in the last 5 years<br />2005-2011<br />
    56. 56. AGRIS Contributions 5 years<br />
    57. 57. Thank You!<br />http://www.ciard.net<br />http://ring.ciard.net<br />http://aims.fao.org<br />http://agris.fao.org<br />

    ×