Assess Criminal Recidivism Risk
About a fifth of all convicted criminal offenders in Switzerland reoffend after serving their sentence. In order to avoid future crimes and support former convicts in their reintegration into society, this rate needs to be lowered. In other words, law enforcement agencies need to be able to predict who is likely to commit another criminal offense.
Law enforcement agencies have access to various information on convicted criminals incl. reoffenders like their criminal records, disciplinary infractions while imprisoned etc.). Using Machine Learning the knowledge contained within these datasets can be extracted. Law enforcement agents can build and apply ML models to future data to predict if a convicted criminal is likely to reoffend or not. ML models furthermore offer insights on which parameters are crucial for a prediction. This allows law enforcement professionals to decide on the right measures to support people in danger of reoffending to eventually prevent crime.
Modulos AutoML makes the creation of effective ML models easy and accessible to domain experts. The platform and the ML models can furthermore also be deployed on-premise, making them well suited for the analysis of potentially sensitive data.