Born from the wish to make linking tractable, the Link Discovery Framework for Metric Spaces (LIMES) is tailored towards the time-efficient and lossless discovery of links across knowledge bases. LIMES is an extensible declarative framework that encapsulates manifold algorithms dedicated to the processing of structured data of any sort. Built with extensibility and easy integration in mind, LIMES allows implementing applications that integrate, consume and/or generate Linked Data. Within LOD2, it will be used for discovering links between knowledge bases.
This webinar will be presented by the LOD2 Partner: University of Leipzig (ULEI), Germany.
1. Creating Knowledge out of Interlinked Data
LOD2 Webinar . 29.11.2011 . Page 1 http://lod2.eu
2. Creating Knowledge out of Interlinked Data
LOD2 is a large-scale integrating project co-funded by the European
Commission within the FP7 Information and Communication Technologies
Work Programme. This 4-year project comprises leading Linked Open
Data technology researchers, companies, and service providers. Coming
from across 12 countries the partners are coordinated by the Agile
Knowledge Engineering and Semantic Web Research Group at the
University of Leipzig, Germany.
LOD2 will integrate and syndicate Linked Data with existing large-scale
applications. The project shows the benefits in the scenarios of Media and
Publishing, Corporate Data intranets and eGovernment.
http://lod2.eu
LOD2 Webinar . 29.11.2011 . Page 2 http://lod2.eu
3. Creating Knowledge out of Interlinked Data
Once per month the LOD2 webinar series offer a free webinar about
tools and services along the Linked Open Data Life Cycle.
Stay with us and learn more about acquisition, editing, composing,
connected applications – and finally publishing Linked Open Data.
http://lod2.eu
LOD2 Webinar . 29.11.2011 . Page 3 http://lod2.eu
4. Creating Knowledge out of Interlinked Data
LIMES
- Link Discovery Framework for Metric Spaces -
LOD2 Webinar . 20.03.2012 . Page 4 http://lod2.eu
5. Creating Knowledge out of Interlinked Data
Overview
• LIMES in LOD2
• Main Ideas
• Technical Details
• Using LIMES
– The „Geeky“ Approach
– LIMES Interface
– Assisted Linking
LOD2 Webinar . 20.03.2012 . Page 5 http://lod2.eu
6. Creating Knowledge out of Interlinked Data
Inter-
linking/
Fusing
Linked Data Lifecycle
Manual Classifi-
revision/ cation/
authoring Enrichment
Storage/
Linked Data Quality
Querying
Lifecycle Analysis
Evolution /
Extraction Repair
Search/
Browsing/
Exploration
LOD2 Webinar . 20.03.2012 . Page 6 http://lod2.eu
7. Creating Knowledge out of Interlinked Data
LOD2 Services LOD2 Stack UI components
Sindice Sig.ma Browse &
LOD2 Stack Semantic
Structure
Authoring
GovData.eu
eGovernment LOD2 Stack APIs
Portal and components Interlinking API
Exalead Enterprise Search DXX LIMES SILK
Web
Search Enrichment and Repair API
Multi-Domain
Ontology
DL-Learner ORE
LOD2 STACK API Knowledge Base Fusion API
Create
Wolters Kluwer Deutschland Structure SemMF WIQA
LOD2 applied to Media and Publishing Link
Fuse
Knowledge Store API
LOD Cloud hosted on OpenLink's Virtuoso
Query and Browsing capability Virtuoso + MonetDB
LOD Cloud: Access interfaces:
Knowledge Storage Layer
Linked Data SPARQL DUMPS Triplify, D2R
Central LOD2 Services Distributed/Local LOD2 Components
LOD2 Webinar . 20.03.2012 . Page 7 http://lod2.eu
8. Creating Knowledge out of Interlinked Data
Link Discovery
• Characteristics
– Very large data sets
– Complex data sets
• Problems
– Runtime
– Complex Specifications
• Solutions
– Time-efficient computation
– Assistance during configuration
– Machine learning for creating link specifications
LOD2 Webinar . 20.03.2012 . Page 8 http://lod2.eu
9. Creating Knowledge out of Interlinked Data
LIMES
• Declarative Link Discovery Framework
• Tuned towards efficiency and extensibility
• Set-theoretical grammar for specifying links
• Time-efficient mappers for single data types
• Machine learning for detecting link specs
LOD2 Webinar . 20.03.2012 . Page 9 http://lod2.eu
10. Creating Knowledge out of Interlinked Data
Architecture
Machine Learning
LOD2 Webinar . 20.03.2012 . Page 10 http://lod2.eu
11. Creating Knowledge out of Interlinked Data
Workflow
LOD2 Webinar . 20.03.2012 . Page 11 http://lod2.eu
12. Creating Knowledge out of Interlinked Data
Workflow
Hybrid
approach
Time-efficient
Rich grammar mappers
LOD2 Webinar . 20.03.2012 . Page 12 http://lod2.eu
13. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
1. Metadata
2. Source and Target
3. Similarity Measure
4. Acceptance Conditions
5. Review Conditions
6. Execution Mode
7. Output Format
LOD2 Webinar . 20.03.2012 . Page 13 http://lod2.eu
14. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
1. Metadata
2. Source and Target
3. Similarity Measure
4. Acceptance Conditions
5. Review Conditions
6. Execution Mode
7. Output Format
LOD2 Webinar . 20.03.2012 . Page 14 http://lod2.eu
15. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
1. Metadata
2. Source and Target
3. Similarity Measure
4. Acceptance Conditions
5. Review Conditions
6. Execution Mode
7. Output Format
LOD2 Webinar . 20.03.2012 . Page 15 http://lod2.eu
16. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
• Preprocessing functions
– Strings, numerical values
– Data converters
• Similarity Measures Trigram
– String
– Numerical values
lowerCase
label label
LOD2 Webinar . 20.03.2012 . Page 16 http://lod2.eu
17. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
• Operators
– Measure operators
– Spec operators
MAX
Trigram Trigram
label label label name
LOD2 Webinar . 20.03.2012 . Page 17 http://lod2.eu
18. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
• Operators
– Measure operators
OR
– Spec operators
Filter Filter
Trigram Trigram
label label label name
LOD2 Webinar . 20.03.2012 . Page 18 http://lod2.eu
19. Creating Knowledge out of Interlinked Data
LIMES Link Specifications
1. Metadata
2. Source and Target
3. Similarity Measure
4. Acceptance Conditions
5. Review Conditions
6. Execution Mode
7. Output Format
LOD2 Webinar . 20.03.2012 . Page 19 http://lod2.eu
20. Creating Knowledge out of Interlinked Data
Geeky approach: XML
• Task: Link drugs
and ingredients
– Source: Dailymed
– Target: Drugbank
– Features
• Definition of source, target, measures
• Using property chains for linking
• Using preprocessing
LOD2 Webinar . 20.03.2012 . Page 20 http://lod2.eu
21. Creating Knowledge out of Interlinked Data
LIMES Native Interface
• Task: Link drugs
across knowledge bases
– Source: DBpedia
– Target: Drugbank
– Features
• Definition of complex measures
LOD2 Webinar . 20.03.2012 . Page 21 http://lod2.eu
22. Creating Knowledge out of Interlinked Data
LIMES Native Interface
• OR(trigram(x.rdfs:label, y.drugbank:genericName)|0.8,
trigram(x.rdfs:label, y.rdfs:label)|0.8)
OR
Filter Filter
Trigram Trigram
label label label genericName
LOD2 Webinar . 20.03.2012 . Page 22 http://lod2.eu
23. Creating Knowledge out of Interlinked Data
COLANUT
• Task: Link diseases
across knowledge bases
– Source: Diseasome
– Target: Sider
– Features
• Assisted linking
• Definition of complex measures
LOD2 Webinar . 20.03.2012 . Page 23 http://lod2.eu
24. Creating Knowledge out of Interlinked Data
Further Information
• Technical Details
– Requirements: Java 1.6
– License: http://creativecommons.org/licenses/
by-nc-sa/3.0/
• Technical papers
– Axel-Cyrille Ngonga Ngomo: A Time-Efficient Hybrid Approach to Link
Discovery. In: Proceedings of the sixth international workshop on
Ontology Matching, 2011
– Axel-Cyrille Ngonga Ngomo und Klaus Lyko: EAGLE: Efficient Active
Learning of Link Specifications using Genetic Programming. In:
Proceedings of ESWC 2012
– Axel-Cyrille Ngonga Ngomo, Jens Lehmann, Sören Auer und Konrad
Höffner: RAVEN -- Active Learning of Link Specifications. In: Proceedings
of OM@ISWC
LOD2 Webinar . 20.03.2012 . Page 24 http://lod2.eu
25. Creating Knowledge out of Interlinked Data
Credits
Jingle Axel Ngonga
Coordination Thomas Thurner
Martin Kaltenböck
Moderation Martin Kaltenböck
Presented by Axel Ngonga
http://bis.uni-leipzig.de/AxelNgonga
ngonga@informatik.uni-leipzig.de
LOD2 Webinar . 29.11.2011 . Page 25 http://lod2.eu
26. Creating Knowledge out of Interlinked Data
Hope you enjoyed staying with us – if you need more detailed
information, visit us at www.lod2.eu and let us know how we can
improve to meet your expectations!
Don’t forget to register for our next webinar
24.04.2012 – D2R (University of Leipzig)
http://lod2.eu
LOD2 Webinar . 29.11.2011 . Page 26 http://lod2.eu