Towards a Linked Open Data Cloud of Language Resources in the Legal Domain - Law Via Internet
Patricia Martín-Chozas, Elena Montiel-Ponsoda,
Víctor Rodríguez-Doncel
10th October 2018
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Towards a Linked Open Data Cloud of Language Resources in the Legal Domain
1. Building the Legal Knowledge Graph
for Smart Compliance Services in Multilingual Europe
http://lynx-project.eu/
Towards a Linked Open Data
Cloud of Language Resources
in the Legal Domain
Patricia Martín-Chozas, Elena Montiel-Ponsoda,
Víctor Rodríguez-Doncel
10th October 2018
2. OVERVIEW
Lynx Goal: Legal Knowledge Graph (LKG)
What, why and how?
Contribution: Linguistic Linked Legal Open Data cloud
(LLLOD)
Linguistic foundation of the LKG
Current status
First approach of the LLLOD
Future Work
Improvements and next steps
3. LYNX PROJECT http://lynx-project.eu/
European project to build smart services towards compliance
Goal: Legal Knowledge Graph
Figure 1. From Lynx Document of Work
4. WHAT IS A KNOWLEDGE GRAPH?
Knowledge graph:
Set of connected pieces
of information
Figure 2. From Ambiverse website
5. WHY DO WE NEED THE LKG?
Smart Services: multijurisdictional open legal knowledge in
one single platform
Access to multilingual information
Customised recommendations
Alerts
Easier compliance for SMEs and large enterprises
6. WHY DO WE NEED THE LKG?
Smart Services: multijurisdictional open legal knowledge in
one single platform
Access to multilingual information
Customised recommendations
Alerts
Easier compliance for SMEs and large enterprises
Law for everybody
7. HOW DO WE BUILD THE LKG?
Figure 3. From Lynx website
8. HOW DO WE BUILD THE LKG?
Legal Language Resources
Figure 3. From Lynx website
9. LEGAL LANGUAGE RESOURCES
Language resources: structured linguistic data in machine
readable forms
Glossaries
Terminologies
Databases
Thesauri
Dictionaries
Lexicons
10. LEGAL LANGUAGE RESOURCES
Language resources: structured linguistic data in machine
readable forms
Glossaries
Terminologies
Databases
Thesauri
Dictionaries
Lexicons
Legal domain
13. LINGUISTIC LINKED OPEN DATA CLOUD
Legal domain is
underrepresented
Need for a Linguistic Linked
Open Data cloud in the legal
domain
Figure 6. From Linked Open Data website
15. CONTRIBUTION
Identification
• Identification of existing Legal
Language Resources
Creation
• Creation of new Legal
Language Resources
Conversion • Transformation of such resources
into RDF
Linking • Linking with other
resources
16. IDENTIFICATION
36 linguistic datasets
General and legal domain
Various formats
Multiple languages
Different textual typology
Available in RDF 16 datasets (44%)
Available in other formats 6 datasets (17%)
Archived resources 14 datasets (39%)
36 datasets
First results
17. CREATION
Labour Law
Glossary (TBX)
Data Protection
Glossary (TBX)
Industrial Standards
Glossary (TBX)
Labour Law
Corpora (PDF)
Data Protection
Corpora (TXT)
Industrial Standards
Corpora (PDF)
Term Extraction
Resulting language resources
Labour Law
Glossary (ES)
Data Protection
Glossary (EN)
Industrial Standards
Glossary (EN)
102 terms 98 terms 109 terms
21. FUTURE WORK
Further transformation of resources into RDF
Automatically enrich glossaries with:
• Translations
• Definitions
• Usage contexts
Automatically link glossaries with other resources