SlideShare a Scribd company logo
Creating Knowledge out of Interlinked Data




        D3.1.1 - Knowledge Extraction
        D3.2.1 - NLP2RDF + NIF




                                             Sebastian Hellmann
                                                        AKSW, Universität Leipzig
LOD2 Presentation . 02.09.2010 . Page                                http://lod2.eu
Creating Knowledge out of Interlinked Data




D 3.1.1 Knowledge Extraction from Structured Sources




      • Results of the Deliverable (1):
            • Definition of Knowledge Extraction on Wikipedia (Provide an
              easy entry point for interested users)




LOD2 Event . 06.09.2010 . 2Page 2                                  http://lod2.eu
Distinction to Information Extraction and ETL and Ontology Learning
How can we define Knowledge?


                      3
4
Creating Knowledge out of Interlinked Data




D 3.1.1 Knowledge Extraction from Structured Sources




      • Results of the Deliverable (2):
            • Tool Server on http://data.lod2.eu/2011/tools/
            • Survey available at http://tinyurl.com/KETSurvey




LOD2 Event . 06.09.2010 . 5Page 5                                http://lod2.eu
6
7
8
Creating Knowledge out of Interlinked Data




D 3.1.1 Knowledge Extraction from Structured Sources




      • Integration of Knowledge Extraction tools is done over the
          format RDF and reusing vocabularies




LOD2 Event . 06.09.2010 . 9Page 9                                http://lod2.eu
Creating Knowledge out of Interlinked Data




D 3.2.1 NLP2RDF + NIF

    NLP2RDF + NIF

      • Vision: Integration of NLP tools based on an Ontological
          Interface
           • String Ontology
           • Structured Sentence Ontology
           • OLiA
           • POWLA

      • NLP Interchange Format(NIF) allows NLP tools to interoperate
      • NLP2RDF provides a reference implementation



LOD2 Event . 06.09.2010 . 10
                           Page 10                                 http://lod2.eu
Creating Knowledge out of Interlinked Data




D 3.2.1 NLP2RDF + NIF



      • Basic idea: adress Strings with URIs
      • Use the expressiveness and flexibility of RDF to add arbitrary
          annotations

      • Several formats:
            • NIF-OWL
            • NIF-RDFa
            • NIF-POWLA




LOD2 Event . 06.09.2010 . 11
                           Page 11                                http://lod2.eu
Creating Knowledge out of Interlinked Data




D 3.2.1 NLP2RDF + NIF




      •




LOD2 Event . 06.09.2010 . 12
                           Page 12                            http://lod2.eu
Creating Knowledge out of Interlinked Data




D 3.2.1 NLP2RDF + NIF

      • Due End of April
      • Iterations: Design and Implementations followed by Feedback
      • All information on http://aksw.org/Projects/NIF
      • First Round (NIF Web Services):
            •   OpenCalais
            •   DBpedia Spotlight
            •   Gate
            •   Stanford Parser (POS Tags)
            •   Lemmatizer


LOD2 Event . 06.09.2010 . 13
                           Page 13                            http://lod2.eu
Creating Knowledge out of Interlinked Data




        Thank you for your
        attention!




LOD2 Presentation . 02.09.2010 . Page                   http://lod2.eu

More Related Content

PDF
NIF - NLP Interchange Format
PPTX
Fremeatfeisgiltt2015 fremelinkeddatalocalisers-150603090934-lva1-app6891
PPTX
Freme at feisgiltt 2015 freme & linked data & localisers
PDF
LDCache - a cache for linked data-driven web applications
PDF
Nobel Prizes as Linked Open Data
PDF
:me owl:sameAs flickr:33669349@N00 .
PDF
NIF 2.0 draft for Pisa
NIF - NLP Interchange Format
Fremeatfeisgiltt2015 fremelinkeddatalocalisers-150603090934-lva1-app6891
Freme at feisgiltt 2015 freme & linked data & localisers
LDCache - a cache for linked data-driven web applications
Nobel Prizes as Linked Open Data
:me owl:sameAs flickr:33669349@N00 .
NIF 2.0 draft for Pisa

Viewers also liked (7)

PPTX
Linked Data for Information Extraction Challenge - Tasks and Results @ ISWC 2014
ODP
Information Extraction from the Web - In today's web
PPTX
Web Information Extraction for the Database Research Domain
PPTX
Textmining Information Extraction
PPTX
Knowledge Extraction from Social Media
PPT
Information extraction 1
ODP
Information Extraction from the Web - Algorithms and Tools
Linked Data for Information Extraction Challenge - Tasks and Results @ ISWC 2014
Information Extraction from the Web - In today's web
Web Information Extraction for the Database Research Domain
Textmining Information Extraction
Knowledge Extraction from Social Media
Information extraction 1
Information Extraction from the Web - Algorithms and Tools
Ad

Similar to LOD2: State of Play WP3B - Knowledge Extraction, NLP2RDF + NIF (20)

PDF
LOD2: State of Play WP3A - Knowledge Base Creation, Enrichment and Repair
ODP
Linked Data for Abbreviations and Segmentation
PDF
LOD2 Webinar Series Classification and Quality Analysis with DL Learner and ORE
PPT
LOD2 Webinar Series: D2R and Sparqlify
PDF
LOD2 Webinar Series: Zemanta / Open refine
PDF
LOD2: State of Play WP5 - Linked Data Visualization, Browsing and Authoring
PDF
LOD2 Plenary Vienna 2012: WP3 - Knowledge Base Creation, Enrichment and Repair
PPT
LOD2 Webinar Series: LOD2 in information and publishing industry
ODP
NIF - Version 1.0 - 2011/10/23
PDF
LOD2: State of Play WP9: Use Case Open Government Data
PDF
LOD2 Webinar Series: DBpedia Spotlight
ODP
Integrating NLP using Linked Data
PDF
OntoWiki Application Framework & Erfurt API
ODP
NIF 2.0 Phd thesis intermediate report
PPT
ODP
NIF 2.0 Tutorial: Content Analysis and the Semantic Web
PDF
Linked Data in Linguistics for NLP and Web Annotation
PDF
Free Webinar: LOD2 Stack - 1st release
LOD2: State of Play WP3A - Knowledge Base Creation, Enrichment and Repair
Linked Data for Abbreviations and Segmentation
LOD2 Webinar Series Classification and Quality Analysis with DL Learner and ORE
LOD2 Webinar Series: D2R and Sparqlify
LOD2 Webinar Series: Zemanta / Open refine
LOD2: State of Play WP5 - Linked Data Visualization, Browsing and Authoring
LOD2 Plenary Vienna 2012: WP3 - Knowledge Base Creation, Enrichment and Repair
LOD2 Webinar Series: LOD2 in information and publishing industry
NIF - Version 1.0 - 2011/10/23
LOD2: State of Play WP9: Use Case Open Government Data
LOD2 Webinar Series: DBpedia Spotlight
Integrating NLP using Linked Data
OntoWiki Application Framework & Erfurt API
NIF 2.0 Phd thesis intermediate report
NIF 2.0 Tutorial: Content Analysis and the Semantic Web
Linked Data in Linguistics for NLP and Web Annotation
Free Webinar: LOD2 Stack - 1st release
Ad

More from LOD2 Creating Knowledge out of Interlinked Data (19)

PPTX
LOD2 Webinar Series: 3rd relase of the Stack
PDF
LOD2 Webinar Series: Virtuoso 7
PDF
LOD2 Webinar Series: publicdata.eu and CKAN
PDF
LOD2 General Presentation 2012
PPT
LOD2 Webinar Series: PoolParty
PDF
LOD2 Plenary Vienna 2012: WP12 - Project Management
PPT
LOD2 Plenary Vienna 2012: WP10 - Training, Dissemination, Community Building,...
PDF
LOD2 Plenary Vienna 2012: WP9A - LOD for a Distributed Marketplace for Public...
ODP
LOD2 Plenary Vienna 2012: WP9 publicdata.eu – Publishing Governmental Informa...
PPTX
LOD2 Plenary Vienna 2012: WP8: Linked Open Data for Enterprise Data Web
PPT
LOD2 Plenary Vienna 2012: WP7 - Linked Open Data for Media and Publishing
PPTX
LOD2 Plenary Vienna 2012: WP6 - Interfaces, Integration & LOD2 Stack
PDF
LOD2 Plenary Vienna 2012: WP5 - Linked Data Browsing, Visualization and Autho...
PDF
LOD2 Plenary Vienna 2012: WP4 - Reuse, Interlinking and Knowledge Fusion
PPTX
LOD2 Plenary Vienna 2012: WP2 - Storing and Querying Very Large Knowledge Bases
PDF
ODP
LOD2 webinar series: Virtuoso by OpenLink Software
LOD2 Webinar Series: 3rd relase of the Stack
LOD2 Webinar Series: Virtuoso 7
LOD2 Webinar Series: publicdata.eu and CKAN
LOD2 General Presentation 2012
LOD2 Webinar Series: PoolParty
LOD2 Plenary Vienna 2012: WP12 - Project Management
LOD2 Plenary Vienna 2012: WP10 - Training, Dissemination, Community Building,...
LOD2 Plenary Vienna 2012: WP9A - LOD for a Distributed Marketplace for Public...
LOD2 Plenary Vienna 2012: WP9 publicdata.eu – Publishing Governmental Informa...
LOD2 Plenary Vienna 2012: WP8: Linked Open Data for Enterprise Data Web
LOD2 Plenary Vienna 2012: WP7 - Linked Open Data for Media and Publishing
LOD2 Plenary Vienna 2012: WP6 - Interfaces, Integration & LOD2 Stack
LOD2 Plenary Vienna 2012: WP5 - Linked Data Browsing, Visualization and Autho...
LOD2 Plenary Vienna 2012: WP4 - Reuse, Interlinking and Knowledge Fusion
LOD2 Plenary Vienna 2012: WP2 - Storing and Querying Very Large Knowledge Bases
LOD2 webinar series: Virtuoso by OpenLink Software

LOD2: State of Play WP3B - Knowledge Extraction, NLP2RDF + NIF

  • 1. Creating Knowledge out of Interlinked Data D3.1.1 - Knowledge Extraction D3.2.1 - NLP2RDF + NIF Sebastian Hellmann AKSW, Universität Leipzig LOD2 Presentation . 02.09.2010 . Page http://lod2.eu
  • 2. Creating Knowledge out of Interlinked Data D 3.1.1 Knowledge Extraction from Structured Sources • Results of the Deliverable (1): • Definition of Knowledge Extraction on Wikipedia (Provide an easy entry point for interested users) LOD2 Event . 06.09.2010 . 2Page 2 http://lod2.eu
  • 3. Distinction to Information Extraction and ETL and Ontology Learning How can we define Knowledge? 3
  • 4. 4
  • 5. Creating Knowledge out of Interlinked Data D 3.1.1 Knowledge Extraction from Structured Sources • Results of the Deliverable (2): • Tool Server on http://data.lod2.eu/2011/tools/ • Survey available at http://tinyurl.com/KETSurvey LOD2 Event . 06.09.2010 . 5Page 5 http://lod2.eu
  • 6. 6
  • 7. 7
  • 8. 8
  • 9. Creating Knowledge out of Interlinked Data D 3.1.1 Knowledge Extraction from Structured Sources • Integration of Knowledge Extraction tools is done over the format RDF and reusing vocabularies LOD2 Event . 06.09.2010 . 9Page 9 http://lod2.eu
  • 10. Creating Knowledge out of Interlinked Data D 3.2.1 NLP2RDF + NIF NLP2RDF + NIF • Vision: Integration of NLP tools based on an Ontological Interface • String Ontology • Structured Sentence Ontology • OLiA • POWLA • NLP Interchange Format(NIF) allows NLP tools to interoperate • NLP2RDF provides a reference implementation LOD2 Event . 06.09.2010 . 10 Page 10 http://lod2.eu
  • 11. Creating Knowledge out of Interlinked Data D 3.2.1 NLP2RDF + NIF • Basic idea: adress Strings with URIs • Use the expressiveness and flexibility of RDF to add arbitrary annotations • Several formats: • NIF-OWL • NIF-RDFa • NIF-POWLA LOD2 Event . 06.09.2010 . 11 Page 11 http://lod2.eu
  • 12. Creating Knowledge out of Interlinked Data D 3.2.1 NLP2RDF + NIF • LOD2 Event . 06.09.2010 . 12 Page 12 http://lod2.eu
  • 13. Creating Knowledge out of Interlinked Data D 3.2.1 NLP2RDF + NIF • Due End of April • Iterations: Design and Implementations followed by Feedback • All information on http://aksw.org/Projects/NIF • First Round (NIF Web Services): • OpenCalais • DBpedia Spotlight • Gate • Stanford Parser (POS Tags) • Lemmatizer LOD2 Event . 06.09.2010 . 13 Page 13 http://lod2.eu
  • 14. Creating Knowledge out of Interlinked Data Thank you for your attention! LOD2 Presentation . 02.09.2010 . Page http://lod2.eu