CJMS Dataset- Cross Jurisdictional legal text for modern slavery in supply chain context classification

We compiled a new dataset of judicial documents (court opinions and case
dockets) that involve modern slavery in a supply chain context focused on two jurisdictions, the USA and India, to
incorporate diverse legal vocabularies and settings:
• The USA Court Opinions from CourtListener: 1,725 judicial opinions were programmatically retrieved using
the CourtListener API2. Queries targeted documents containing keywords indicative of modern slavery (e.g.,
"forced labor", "forced labour", "human trafficking", "involuntary servitude", "child labor", "child labour",
"modern slavery") specifically within the context of supply chains, often requiring co-occurrence with the supply
chain-related terminology. CourtListener is (www.courtlistener.com), provides access to a comprehensive set of
federal cases from 1926 onward of United States Supreme Court (Adusumilli et al., 2022). The site is sponsored
by the non-profit Free Law Project.
• Indian Court Cases from Legal Information Institute of India (LIIofIndia)3: 436 legal cases documents were
sourced from the Indian legal database LII of India using custom Python-based web scraping scripts. Search
query formulations were adapted to be equivalent to those used for the USA data, accounting for potential
variations in terminology.

1752 cases negatives for modern slavery
601 cases positive for modern slavery where 311 is indicative of modern slavery in supply chain related to produce of good or services under at least one of this forced labour, child labour, sex trafficking, human trafficking categories.