Optimizing predictive performance of criminal recidivism models using registration data with binary and survival outcomes

Abstract

In a recidivism prediction context, there is no consensus on which modeling strategy should be followed for obtaining an optimal prediction model. In previous papers, a range of statistical and machine learning techniques were benchmarked on recidivism data with a binary outcome. However, two important tree ensemble methods, namely gradient boosting and random forests were not extensively evaluated. In this paper, we further explore the modeling potential of these techniques in the binary outcome criminal prediction context. Additionally, we explore the predictive potential of classical statistical and machine learning methods for censored time-to-event data. A range of statistical manually specified statistical and (semi-)automatic machine learning models is fitted on Dutch recidivism data, both for the binary outcome case and censored outcome case. To enhance generalizability of results, the same models are applied to two historical American data sets, the North Carolina prison data. For all datasets, (semi-) automatic modeling in the binary case seems to provide no improvement over an appropriately manually specified traditional statistical model. There is however evidence of slightly improved performance of gradient boosting in survival data. Results on the reconviction data from two sources suggest that both statistical and machine learning should be tried out for obtaining an optimal model. Even if a flexible black-box model does not improve upon the predictions of a manually specified model, it can serve as a test whether important interactions are missing or other misspecification of the model are present and can thus provide more security in the modeling process.

Integrating gut microbiome and host immune markers to understand the pathogenesis of Clostridioides difficile infection

“…Classical logistic regression is one of the most common machine learning models in medicine. Yet, it fails to solve non-linear problems where there are multiple or non-linear decision boundaries 64 . Furthermore, the log odds scale in LR is hard to interpret 65 .…”

Machine Learning and Criminal Justice: A Systematic Review of Advanced Methodology for Recidivism Risk Prediction

“…The authors used a dataset from the Minnesota Screening Tool Assessing Recidivism Risk (MnSTARR), which assesses the risk of five different types of recidivism, taking advantage of the Minnesota Sex Offender Screening Tool-3 (MnSOST-3), used to analyse sexual recidivism risk for Minnesota sex offenders. Tollenaar and colleagues in two different studies used the StatRec scale with static information from the Dutch Offender Index (DOI) [ 31 , 32 ]. In both studies, the recidivism prediction is divided into three categories: general, criminal and violent recidivism.…”

Integrating Gut Microbiota and Host Immune Markers for Highly Accurate Diagnosis of Clostridioides difficile Infection

“…Classical logistic regression, is one of the most common machine learning models in medicine 75 . The main drawback of LR is its failure to solve non-linear problems and it underperforms where there are multiple or non-linear decision boundaries 76 . Furthermore, the log odds scale in LR is hard to interpret 77 .…”