Code - How to Use Large Language Models for Empirical Legal Research

Code - How to Use Large Language Models in Empirical Legal Research


The following code is used in my work in progress, How to Use Large Language Models in Empirical Legal Research. The files should be run in Python (.py).

I’m also happy to provide the datasets used in the article—please email me at jonchoi@law.usc.edu, if interested. The human-generated data were created for Nina Mendelson’s original paper, Change, Creation, and Unpredictability in Statutory Interpretation: Interpretive Canon Use in the Roberts Court's First Decade, and should be credited accordingly. I only analyzed a small subset of Prof. Mendelson’s original dataset, please contact her for the complete data.

llm_canon_project.py

llm_canon_project_statistics.py

This page was last updated on August 10, 2023.