U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Machine learning risk assessments in criminal justice settings

NCJ Number
306018
Author(s)
Richard Berk
Date Published
2019
Length
8 pages
Annotation

This book attempts increase the accessibility of the most recent work on forecasts of re-offending by individuals already in criminal justice custody.

Abstract

In this book, the author attempts to increase the accessibility of the most recent work on forecasts of re-offending by individuals already in criminal justice custody. The book’s target audience is graduate students and researchers in the social sciences, and data analysts in criminal justice agencies. The author assumes the reader has a working knowledge of the generalized linear model, and notes that all empirical examples were constructed using the R programming language, which is an open access software. The goal of this work, aside from aggregating research, is to use machine learning statistical procedures trained on very large datasets, and emphasize the maximization of forecasting accuracy.