Using Data Science to Study the U.S. Federal District Courts

25 January 2024, 10am

This talk introduces the Systematic Content Analysis of Litigation Events (SCALES) project, a multi-year, multi-institution effort to create a free, publicly accessible archive of federal court records and data from the U.S. district courts. Working with a corpus of millions of docket sheet entries and court documents, we built a classification pipeline to label important litigation events in both civil and criminal cases to identify cases’ pathways and outcomes. We also built a custom entity disambiguation approach to identify judges, lawyers, law firms, and party types (e.g. individual vs. business vs. government). The talk will introduce the SCALES database and present several in-progress research projects that use the SCALES data and/or were inspired by the SCALES approach.