From Text to Structure: Using Large Language Models to Support the Development of Legal Expert Systems

TL;DR

This paper explores using large language models like GPT-4 to automatically extract structured legal representations from legislation, aiming to streamline the development of transparent, rule-based expert systems in law.

Contribution

It demonstrates that LLMs can generate legal pathways comparable to manual ones, reducing the time and effort needed for formalizing legislation in expert systems.

Findings

60% of generated pathways rated as equivalent or better than manual ones

LLMs can effectively support the encoding of legislative texts

Potential to significantly reduce development time for legal expert systems

Abstract

Encoding legislative text in a formal representation is an important prerequisite to different tasks in the field of AI & Law. For example, rule-based expert systems focused on legislation can support laypeople in understanding how legislation applies to them and provide them with helpful context and information. However, the process of analyzing legislation and other sources to encode it in the desired formal representation can be time-consuming and represents a bottleneck in the development of such systems. Here, we investigate to what degree large language models (LLMs), such as GPT-4, are able to automatically extract structured representations from legislation. We use LLMs to create pathways from legislation, according to the JusticeBot methodology for legal decision support systems, evaluate the pathways and compare them to manually created pathways. The results are promising, with 60% of generated pathways being rated as equivalent or better than manually created ones in a blind comparison. The approach suggests a promising path to leverage the capabilities of LLMs to ease the costly development of systems based on symbolic approaches that are transparent and explainable.