Understanding Judicial Bias Through Machine Learning

I am working with Anjanette Raymond at the Ostrom Workshop at Indiana University on a project to use data mining techniques to help us learn more about the nature of judicial bias. Currently the study is in the exploratory phase. Following past literature I am using random forests to predict the outcomes of supreme court cases using the supreme court database (SCBD). After confirming the predictive power of this method, I will begin engineering extralegal features (race of plaintif(s), gender of plaintif(s), etc…) while omiitting legal features (issues, circuit of origin, etc…). If the method still carries predictive power, I will begin looking at what features, or combinations of features, might serve as predictors for case outcomes.

The SCBD is just the most easily accessable dataset for court cases. Ideally, this study will focus just as much on local-scale cases. Of particular interest are traffic violations, as we suspect that many areas might store this data along with the demographic information from drivers licenses, allowing us to avoid the arduous process of manually finding and entering feature data.