compas dataset. From ProPublica: across the nation, judges, probation and parole officers are increasingly using algorithms to assess a criminal defendant’s likelihood to re-offend.
data(compas)A data frame with 6172 rows and 7 variables:
The original source of data is https://www.propublica.org/datastore/dataset/compas-recidivism-risk-score-data-and-analysis. Modified data used here comes from https://www.kaggle.com/danofer/compass/ (probublicaCompassRecidivism_data_fairml.csv)
factor, 1/0 for future recidivism or no recidivism. Models should predict this values
numeric, number of priors
factor, 1/0 for age above 45 years or not
factor, 1/0 for age below 25 years or not
factor, 1/0 for having recorded misdemeanor(s) or not
factor, Caucasian, African American, Asian, Hispanic, Native American or Other
factor, female/male for gender