justiceBRD
Justice Beyond Reasonable Doubt
A computational model designed to reduce wrongful convictions by providing a rigorous framework for mathematically determining guilt based on evidence.
To access our model, view our training and test cases, or run your own cases through our model, please click the 'Get started' button above, or click register (free), or use the 'Sign in/Register' link at the top of the page.
See our Quick Start guide to get going fast.
Wrongfully convicting a person of any crime is an appalling justice failure, with unarguably the most profound injustice being a wrongful conviction for the crime of murder.
We developed this machine-learning model as a tool for assessing trial evidence based on our knowledge of 30 historical murder trials.
A full account of how it works is given in our paper Machine learning for determining accurate outcomes in criminal trials recently published in the peer reviewed journal Law, Probability and Risk.
We are now opening up the model so that those with a fuller knowledge of the cases we used, or with knowledge of new cases, can test the model themselves. Only after a prolonged and thorough testing of the model can its predictive power become known and accepted, and we welcome your contribution to this process.
“It is better to let ten guilty men go free than to wrongly incarcerate one innocent man.”