Research map / Learning theory and algorithms / Fairness
WCLD: Curated Large Dataset of Criminal Cases from Wisconsin Circuit Courts
Elliott Ash, Naman Goel, Nianyun Li, Claudia Marangon, Peiyao Sun
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use XGBoost instead of logistic regression | recidivism classifier · WCLD 180-day cut-off, overall | accuracy: 0.6461 -> 0.6588 | No clear effect |
| use XGBoost instead of logistic regression | recidivism classifier · WCLD 180-day cut-off, overall | AUC: 0.6822 -> 0.7039 | Improved |
| use XGBoost instead of logistic regression | recidivism classifier · WCLD 180-day cut-off, overall | FPR: 0.1532 -> 0.2244 | Hurt |
| use XGBoost instead of logistic regression | recidivism classifier · WCLD 180-day cut-off, overall | FNR: 0.6282 -> 0.5008 | Improved |
| use XGBoost instead of logistic regression | recidivism classifier · WCLD 180-day cut-off, Native American vs Hispanic | PR gap between groups: 0.4800 vs 0.2910 (XGBoost); 0.3761 vs 0.1991 (LR) | Hurt |
| evaluate Native American group vs Hispanic group | XGBoost · WCLD 180-day cut-off | FPR: 0.3272 vs 0.2016 | Hurt |
| evaluate African American group vs Caucasian group | XGBoost · WCLD 180-day cut-off | FPR: 0.2159 -> 0.2454 | Hurt |
| evaluate African American group vs Caucasian group | XGBoost · WCLD 180-day cut-off | FNR: 0.5113 -> 0.4792 | Improved |
| change recidivism label from 180-day to 2-year sentence cut-off | XGBoost · WCLD overall | AUC: 0.7039 -> 0.7018 | No clear effect |
| change recidivism label from 180-day to 2-year sentence cut-off | logistic regression · WCLD overall | FNR: 0.6282 -> 0.6429 | No clear effect |
| change recidivism label from 180-day to 2-year sentence cut-off | XGBoost · WCLD Native American group | FPR: 0.3272 -> 0.3223 | No clear effect |
Closest papers by meaning
- Combining Machine Learning and Queueing Theory for Data-Driven Incarceration-Diversion Program Management (AAAI 2024 · similarity 0.83)
- Evaluating Pre-trial Programs Using Interpretable Machine Learning Matching Algorithms for Causal Inference (AAAI 2024 · similarity 0.81)
- Fairness under Covariate Shift: Improving Fairness-Accuracy Tradeoff with Few Unlabeled Test Samples (AAAI 2024 · similarity 0.80)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · similarity 0.79)
- Fairness without Demographics through Shared Latent Space-Based Debiasing (AAAI 2024 · similarity 0.79)
- Lawma: The Power of Specialization for Legal Annotation (ICLR 2025 · similarity 0.79)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
