What is it about?
We compare the relative performance of a traditional regression method (logistic regression) and three machine learning techniques (classification and decision tree, chi-squared automatic interaction detection, and neural networks) to determine which approach works better in predicting prison inmates who engage in misconduct.
Featured Image
Why is it important?
In the U.S., prison administrators often rely on risk assessment instruments to place and supervise inmates and to plan and allocate resources. Therefore, any improvement in the accuracy performance of risk assessment instruments is likely to result in significant benefits for offender classification and rehabilitation, management systems, and public safety. This is arguably the first study to explore whether machine learning techniques perform better than traditional regression method in predicting inmate misconduct.
Perspectives
Machine learning techniques have their advantages and disadvantages but they have become a relevant and key method in developing risk assessment instruments particularly with the rise in big data. More research examining the utility of machine learning techniques to solve problems in the field of criminology is warranted.
Dr. Fawn T. Ngo
University of South Florida Sarasota-Manatee and New College of Florida Official Bookstore
Read the Original
This page is a summary of: Assessing the Predictive Utility of Logistic Regression, Classification and Regression Tree, Chi-Squared Automatic Interaction Detection, and Neural Network Models in Predicting Inmate Misconduct, American Journal of Criminal Justice, May 2014, Springer Science + Business Media,
DOI: 10.1007/s12103-014-9246-6.
You can read the full text:
Resources
Contributors
The following have contributed to this page
