机器学习对枪击受害风险的预测足以助力预防枪击
Machine Learning Can Predict Shooting Victimization Well Enough to Help Prevent It
Review of Economics and Statistics · 2024 · [{"name": "Sara Heller", "affiliation": ["University of Michigan"]}, {"name": "Benjamin Jakubowski", "affiliation": ["Franklin University", "New York University"]}, {"name": "Zubin Jelveh", "affiliation": ["University of Maryland, College Park"]}, {"name": "Max Kapustin", "affiliation": ["Cornell University"]}]
中文摘要
我们使用芝加哥警方数据训练机器学习模型，预测个人在未来18个月内遭受枪击的风险。样本外预测准确率高得惊人。使用警方数据的一个核心隐忧是将偏误“固化”进模型，即在行为相同的条件下，高估更可能与警方接触的群体所面临的风险。然而，我们的预测能够准确还原不同人口群体的风险。出于法律、伦理和实践方面的障碍，不应利用受害风险预测来确定执法对象。但利用此类预测来确定社会服务对象，既可以提高干预措施减少枪击事件的潜力，也可以增强检测此类减少效应的可用统计功效。
Abstract
Abstract Using Chicago police data, we train a machine learning model to predict the risk of being shot in the next 18 months. Out-of-sample accuracy is strikingly high. A central concern with using police data is “baking in” bias, or overestimating risk for groups likelier to interact with police conditional on behavior. Our predictions, however, accurately recover risk across demographic groups. Legal, ethical, and practical barriers should prevent using victimization predictions to target law enforcement. But using them to target social services could increase both the potential for interventions to reduce shootings and the available statistical power to detect those reductions.