![27ggg
Advanced algorithms for extracting data
Extracting
concepts not
literally
mentioned in
the text
Source Text
The tablets shall be taken with
liquid, and should not be crushed or
chewed. Ontopharmanax may be
taken with or without food
Extracted data
RoutesOfAdministration
{
“routes”: [“OralUse”]
}
Source Text
Ontopharmanax is for
subcutaneous injection only and
shall not be used for intramuscular
injection.
Extracted data
RoutesOfAdministration
{
“routes”: [
“SubcutaneousInjection”
]}
Dealing with
negations.](https://image.slidesharecdn.com/lj5ddz64qcosl31w4dl7-signature-8036fbb5b78c2cc1acdc72f9f7af5fc1f451a20e58753a03b5ae4b4dc5f6f76e-poli-170919155800/85/Session-1-1-linked-data-applied-a-field-report-from-the-netherlands-27-320.jpg)
Session 1.1 linked data applied: a field report from the netherlands
- 1. 1ggg Linked Data: A Commercial Outlook Our LD-Portfolio in 2017 and a perspective on the future Jan E. Voskuil SEMANTiCS 2017 Amsterdam
- 2. 2ggg
- 3. 3ggg 2012 2013 2014 2015 2016 2017 Revenue from Linked Data work SEMANTiCS 2017 3
- 4. 4ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 4
- 5. 5ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 5 Kennisnet supports schools with IT. Provides a basic IT- infrastructure and shares knowledge. Maintains several standards for information exchange.
- 6. 6ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 6 Publication office for the Dutch government organisations. Maintains value lists that have a legal status, e.g., the list of municipalities.
- 7. 7ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 7 Wolters Kluwer Legal & Regulatory offers information, software and tools for legal professionals. Business vocabularies for knowledge management and search.
- 8. 8ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 8 Alliander is an energy network company. We are bringing an open and sustainable energy market closer to the consumer. Maintains several large business vocabularies and large amounts of technical documents.
- 9. 9ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 9 The Netherlands Institute for Sound and Vision is a cultural-historical organization. It collects, preserves and opens the audiovisual heritage. Maintains several large thesauri.
- 10. 10ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 10 The Netherlands’ Cadastre, Land Registry and Mapping Agency – in short Kadaster – collects and registers administrative and spatial data on property and the rights involved. Publishes Linked Data on a large scale.
- 11. 11ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • Wolters Kluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 11 The Dutch police uses Linked Data-technologies to collate data from disparate sources.
- 12. 12ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • WoltersKluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 12 CROW is a non-profit knowledge partner in road construction. CROW maintains thesauri and a large knowledge base.
- 13. 13ggg Our LD-Portfolio: A Snapshot • Kennisnet • KOOP • WoltersKluwer NL • Alliander • Beeld en Geluid • Kadaster • National Police • CROW • OntoPharma SEMANTiCS 2017 13 OntoPharma, a spin-off of Taxonic, delivers solutions for a Data First-approach in the pharma sector. Solutions include automated data extraction, management of reference data and product data, and structured authoring.
- 15. 15ggg Use Cases SEMANTiCS 2017 15 Vocabulary- oriented Data-oriented Documents and data
- 16. 16ggg Use Cases SEMANTiCS 2017 16 Business vocabularies Reference Data Management Collating business concepts across vocabularies Vocabulary- oriented Data-oriented Documents and data
- 17. 17ggg Use Cases SEMANTiCS 2017 17 Business vocabularies Reference Data Management Collating business concepts across vocabularies Vocabulary- oriented Data-oriented Documents and data Publishing interoperable data Publishing datamodels Managing datasets
- 18. 18ggg Use Cases SEMANTiCS 2017 18 Business vocabularies Reference Data Management Collating business concepts across vocabularies Vocabulary- oriented Data-oriented Documents and data Publishing interoperable data Publishing datamodels Managing datasets Semantic ECM Tagging Extraction Indexing Content Classification Structured Authoring
- 19. 19ggg Kennisnet KOOP WoltersKluwerNL Alliander BeeldenGeluid Kadaster Police CROW OntoPharma Managing metadata sets Business vocabularies Publishing datamodels Publishing interoperable data Collating business concepts across vocabularies Reference Data Management Managing datasets Semantic ECM
- 20. 20ggg Business vocabularies Publishing interoperable data Managing datasets Managing metadata sets Semantic ECM Reference Data Management Collating business concepts across vocabularies Publishing datamodels Relative contribution 2017 SEMANTiCS 2017 20
- 21. 21ggg Business vocabularies Publishing interoperable data Managing datasets Managing metadata sets Semantic ECM Reference Data Management Collating business concepts across vocabularies Publishing datamodels Expectations of Relative Growth SEMANTiCS 2017 21
- 22. 22ggg Collating concepts across vocabularies Kennisnet • Develops standards for information exchange in the field of education • Overlapping semantics • Same concepts in different models • Specialized tooling • TopQuadrant Enterprise Vocabulary Network • Custom concept browsers for visualizing semantic overlap
- 23. 23ggg
- 24. 24ggg > The browser shows info about the term “Samengestelde groep” > This term is defined in EDEXML, a data exchange standard
- 25. 25ggg > The browser shows info about the term “Samengestelde groep” > This term is defined in EDEXML, a data exchange standard > Also occurs in UWRL and ECKID > The browser shows a detailed comparison > Similarities and differences made visible > High level of automation
- 26. 26ggg The potential for Semantic ECM Data First • From documents to data • Regulatory requirements • Data extraction as a service • Future outlook: create data and documents in concert
- 27. 27ggg Advanced algorithms for extracting data Extracting concepts not literally mentioned in the text Source Text The tablets shall be taken with liquid, and should not be crushed or chewed. Ontopharmanax may be taken with or without food Extracted data RoutesOfAdministration { “routes”: [“OralUse”] } Source Text Ontopharmanax is for subcutaneous injection only and shall not be used for intramuscular injection. Extracted data RoutesOfAdministration { “routes”: [ “SubcutaneousInjection” ]} Dealing with negations.
- 28. 28ggg The potential for Semantic ECM Serve information in the field • Operatives need specific information • Buried in documents • Semantic Search • Auto-tagging, extraction, indexation • Future outlook: template- based authoring
- 29. 29ggg Semantic ECM • Semantic ECM is transformational • Affects many processes in many ways • Authoring • Redacting • Retrieval • Difficulty is structurally underestimated • Complexity • Need for solid know-how and the right tools • Many projects start and then fail miserably • Still much evangelizing needed! SEMANTiCS 2017 29
- 30. 30ggg Business vocabularies Thesauri & metadata Data publishing & management Semantic ECM Concept Computing Evolving Use Cases SEMANTiCS 2017 30
- 31. 31ggg
- 32. 32ggg • Open Source • Created by Taxonic, with Kadaster • RML standard for mapping & transformation • From relational to RDF • From any format to RDF • First release now available from Github! • https://carml.gitlab.io/plain-html/