Automated detection and resolution of legal cross references: Approach and a study of Luxembourg's legislation

Abstract

When elaborating compliance requirements, analysts need to follow the cross references in the underlying legal texts and consider the additional information in the cited provisions. To enable easier navigation and handling of cross references, automation is necessary for recognizing the natural language patterns used in cross reference expressions (cross reference detection), and for interpreting these expressions and linking them to the target provisions (cross reference resolution). In this paper, we propose a solution for automated detection and resolution of legal cross references. We ground our work on Luxembourg's legislative texts, both for studying the natural language patterns in cross reference expressions and for evaluating the accuracy and scalability of our solution.

An automated framework for detection and resolution of cross references in legal texts

“…The structure of legislative texts in several EU countries, e.g., the Netherlands [11] and Luxembourg [1], are close (but not identical) to the structure depicted in Fig. 3.…”

“…For T4 and T5, the misinterpreted CREs are external references being interpreted as internal ones, thus affecting only precision. The example we gave earlier for case (1) in Section 8.1 ("paragraph 11bis") is the misinterpretation seen in T5. T6 and T7 contain one misinterpreted CRE that is common between the two texts.…”

“…This article is an extension of a previous conference paper [1] published at the 22nd IEEE International Requirements Engineering Conference (RE 2014). The main extensions in this article are: (1) a succinct and yet comprehensive guide for understanding the structure of legal texts, (2) an extended description of the NLP machinery in our approach with more details and a new set of illustrating examples, (3) expansion of our work to the Canadian legal corpus through an investigation of cross references in both the French and English editions of the Personal Health Information Protection Act (PHIPA) [32].…”

“…Cross reference detection is based on the Natural Language (NL) patterns in the CREs. In our previous work [1], we conducted a Grounded Theory (GT) study [8] of Luxembourg's Income Tax Law (circa 2013) [17] with a total of 1223 CREs for identifying the NL patterns used in CREs.…”

“…(3) A health information custodian may disclose personal health information to a prescribed entity under subsection (1) if, (a) the entity has in place practices and procedures to protect the privacy of the individuals whose personal health information it receives and to maintain the confidentiality of the information; In the examples of Fig. 1, the citing provisions depend on the cited ones for definitions and further elaboration.…”