Artificial Neural Networks for Predicting Death Penalty Outcomes

S.T. Karamouzis and D. Wood Harper (USA)


Neural Networks, Forecasting and Prediction, Death Penalty


The United States is the only western democracy that maintains the death penalty. Thirty-eight states and the Federal government prescribe the death penalty. There are 50 Federal offences for which the death penalty can be prescribed. Opposition to the death penalty takes a number of forms from outright abolition to moratoriums during which, it is assumed, the bias of race, class and due process could be corrected. The focus of our research is on the characteristics of cases that determine whether or not the defendant is actually executed. This article presents the development, training, and testing of an Artificial Neural Network for predicting death penalty outcomes. The network was developed as a three-layered perceptron and was trained using the backpropagation principles. For training and testing various experiments were executed. In these experiments, a sample of 1,366 profiles of death row inmates was used. The sample was divided into two sets. The first set of 1,000 profiles was used for training, 66 for cross validation, and the remaining 300 profiles were used for testing.

Important Links:

Go Back