This repository contains code to generate datasets based on two legal domains: the welfare benefit domain and the tort law domain.
Example notebooks are provided for both domains that explain how to use the code to generate the datasets.
For more information regarding the domains, please refer our paper Arguments, Rules and Cases in Law: Resources for aligning learning and reasoning in structured domains.