Semantic Metadata and Compliance Rule Extraction from Legal Texts


CALL: 2017

DOMAIN: IS - Information and Communication Technologies





HOST INSTITUTION: University of Luxembourg

KEYWORDS: Legal Technology, Regulatory Compliance, Software Engineering, Requirements Engineering, Model-Driven Engineering, Machine Learning, Natural Language Processing

START: 2018-01-01

END: 2020-12-31


Submitted Abstract

Legal Technology (LegalTech) is concerned with developing software-based solutions that assist legal professionals, lay individuals, or both in their interactions with legal texts. LegalTech addresses a broad array of topics, including assisted legal reviews, legal search and discovery, legal content management and reporting, legal analytics, and legal compliance analysis.An important prerequisite for many LegalTech applications is to have the structural and semantic properties of legal texts expressed in a machine-analyzable form. These properties constitute the metadata that needs to be recorded alongside the natural-language content of legal texts. Those facets of LegalTech that deal with legal compliance further require precise rules for a software-based operationalization of compliance-related tasks. Given the sheer scale of existing legal corpora, manually creating the required legal metadata and compliance rules is extremely laborious, and requires vast human resources.The SCARLET project will pursue two ambitious but achievable goals: (1) automated extraction of legal metadata from legal texts and, (2) (semi-)automated specification of compliance rules based on the extracted metadata.SCARLET will be conducted in collaboration with a public-service partner, Service central de législation (SCL). SCL is responsible for publishing Luxembourg’s legislation and administrative procedures. In recent years, SCL has invested significantly into making Luxembourg’s legal texts machine-analyzable. The investment is motivated by several important LegalTech use cases, including: (1) providing intelligent legal search facilities, (2) dynamic generation of legal guidance, and (3) rule-based laws with automated compliance analysis capabilities. SCL has already taken valiant steps toward building the infrastructure necessary to support these use cases. However, the country’s legal corpus has not yet been enhanced with the required metadata and rules due to the prohibitively expensive cost of doing so manually. Through automation, SCARLET aims to provide SCL with cost-effective means for constructing the legal metadata and compliance rules that are needed for supporting the above use cases.The research program in SCARLET is highly innovative and expected to yield important academic contributions. At the same time, the project will offer compelling benefits to SCL by directly supporting the LegalTech use cases that SCL is interested in. While SCARLET has competence building in the public sector as its number one priority, the project will also create value by improving citizens’ access to legal services, opening up valorization opportunities in the private sector, and increasing the attractiveness of Luxembourg to businesses via an efficient compliance regime based on explicit and verifiable compliance rules.

This site uses cookies. By continuing to use this site, you agree to the use of cookies for analytics purposes. Find out more in our Privacy Statement