Recidivism of Criminal Offenders
Recidivism of criminal offenders is a difficult topic in criminal justice. If we view the justice system as serving an important rehabilitatory purpose, then recidivism represents indicators for potential improvement in the correctional system. Using a public data-set on two cohorts of inmates released from North Carolina prisons in 1978 and 1980, we build two predictors of whether an individual will offend again in the future. Our first predictor is a one-layer feed-forward neural networks. The second predictor is a random forest.