Interpretable Classification Models for Recidivism Prediction

Abstract

We investigate a long-debated question, which is how to create predictive models of recidivism that are sufficiently accurate, transparent, and interpretable to use for decision-making. This question is complicated as these models are used to support different decisions, from sentencing, to determining release on probation, to allocating preventative social services. Each use case might have an objective other than classification accuracy, such as a desired true positive rate (TPR) or false positive rate (FPR). Each (TPR, FPR) pair is a point on the receiver operator characteristic (ROC) curve. We use popular machine learning methods to create models along the full ROC curve on a wide range of recidivism prediction problems. We show that many methods (SVM, Ridge Regression) produce equally accurate models along the full ROC curve. However, methods that designed for interpretability (CART, C5.0) cannot be tuned to produce models that are accurate and/or interpretable. To handle this shortcoming, we use a new method known as SLIM (Supersparse Linear Integer Models) to produce accurate, transparent, and interpretable models along the full ROC curve. These models can be used for decision-making for many different use cases, since they are just as accurate as the most powerful black-box machine learning models, but completely transparent, and highly interpretable.

45 pages, 17 figures

References in corpus (3)

• Supersparse Linear Integer Models for Optimized Medical Scoring Systems
• Least angle and ${\ell }_1$ penalized regression: A review
• Elaboration on Two Points Raised in ``Classifier Technology and the Illusion of Progress''

Cited by in corpus (14)

• Explaining Explanations in AI
• 'It's Reducing a Human Being to a Percentage'; Perceptions of Justice in Algorithmic Decisions
• Aligning Superhuman AI with Human Behavior: Chess as a Model System
• On the Existence of Simpler Machine Learning Models
• Explainable Machine Learning for Public Policy: Use Cases, Gaps, and Research Directions
• Sensible AI: Re-imagining Interpretability and Explainability using Sensemaking Theory
• Fairness via Explanation Quality: Evaluating Disparities in the Quality of Post hoc Explanations
• Interpretable multiclass classification by MDL-based rule lists
• Optimal randomized classification trees
• EXMOS: Explanatory Model Steering Through Multifaceted Explanations and Data Configurations
• Scientific Inference With Interpretable Machine Learning: Analyzing Models to Learn About Real-World Phenomena
• Interpretability, Then What? Editing Machine Learning Models to Reflect Human Knowledge and Values
• Towards Directive Explanations: Crafting Explainable AI Systems for Actionable Human-AI Interactions
• Trustworthy AI-based crack-tip segmentation using domain-guided explanations