Towards a high-level controlled language for legal sources on the semantic web

Adam Wyner*, Adeline Nazarenko, François Lévy

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingPublished conference contribution

3 Citations (Scopus)

Abstract

Legislation and regulations are required to be structured and augmented in order to make them serviceable on the Internet. However, it is known that it is complex to accurately parse and semantically represent such texts. Controlled languages have been one approach to adjusting to the complexities, where the source text is rewritten in some systematic form. Such an approach is not only costly, but potentially introduces alternative translations which may be undesirable. To navigate between the requirements and complexities, we propose and exemplify a highlevel controlled language that serves as an XML representation for key components of legal content. The language tightly correlates to the source text and also facilitates analysis.

Original languageEnglish
Title of host publicationControlled Natural Language - 5th International Workshop, CNL 2016, Proceedings
PublisherSpringer Verlag
Pages92-101
Number of pages10
Volume9767
ISBN (Print)9783319414973
DOIs
Publication statusPublished - 2016
Event5th International Workshop on Controlled Natural Language, CNL 2016 - Aberdeen, United Kingdom
Duration: 25 Jul 201627 Jul 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9767
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference5th International Workshop on Controlled Natural Language, CNL 2016
Country/TerritoryUnited Kingdom
CityAberdeen
Period25/07/1627/07/16

Keywords

  • Controlled languages
  • Legal rules
  • Natural language simplification
  • Semantic annotation
  • Semantic web

Fingerprint

Dive into the research topics of 'Towards a high-level controlled language for legal sources on the semantic web'. Together they form a unique fingerprint.

Cite this