ABSTRACT
Building upon decades of empirical research and leveraging contamination from other fields, criminology has lately witnessed a significant increase in the use of computational methods. Machine learning could establish itself as the next substantial fracture within the methodological landscape in this field. This chapter will first discuss three of the most prominent trends that contributed to characterizing the increasingly computational nature of criminology: geospatial modeling of crime, networks and crime, and simulation approaches to the analysis of crime. Second, it will quantitatively assess the current state of the art of scholarship at the intersection of crime research and artificial intelligence (AI) exploring a sample of 132 works gathered from the 20 most influential journals and conferences in criminology and AI, according to the Google Scholar rankings. The analysis will show that the number of works is growing, both at the aggregate and the discipline levels, but that authorship is disconnected and that it completely lacks individual collaboration among scholars belonging to the two different fields. In terms of country-level patterns, the United States demonstrates to be the most central hub according to a variety of measures, followed by the United Kingdom and India. Finally, the trends in terms of topics reveal heterogeneous dynamics and the very marginal attention to criminological theory and the ethical dimension of intelligent algorithms deployed for countering, predicting, and reducing crime.
