The platform, based on data from the Public Prosecutor's Office and the SOSAFE mobile application, is aimed at prosecutors, police, and municipalities, and will support both effective criminal prosecution and the design of preventive patrol strategies, providing faster responses in urgent situations.
The project, titled "Scaling and validation of an integrated system for crime prosecution using advanced data analytics," will develop a systems integration and model improvement plan that was presented in the first part of the project FONDEF ID20i10230. This which involved Universidad de Chile, Universidad de los Andes, and Universidad del Bío-Bío, in collaboration with the Public Prosecutor's Office (known as the National Prosecutor's Office) and the company SOSAFE. . research, which uses Artificial Intelligence (AI) for decision support for crime prevention, is co-led by Carla Vairetti, from School of Engineering and Applied Sciences, and won a FONDEF IT grant to continue with the development of its second stage.
The platform for the Public Prosecutor's Office will focus on extracting names of people or places, as well as identifying the modus operandi of crimes, through victims' accounts, using social network analysis and AI, to improve criminal prosecution processes.
"This project improves the prosecution of crime through a platform that integrates various sources of relevant information, using AI and social network analytics. The FONDEF director is Richard Weber (UCH), and my participation is as co-director, together with co-investigators Sebastián Maldonado (UCH) and Freddy Troncoso (UBioBio). Undergraduate students —who will develop their theses— and master's and doctoral students will also participate. In addition, the collaboration of UANDES alumni who worked on the initial phase of the project is also planned," explains Carla Vairetti.
In this new phase of the project two integrated platforms will be created for the prosecution of crimes with a high level of technological maturity demonstrated in real-life situations. "We will expand the crime analysis by incorporating additional sources of information, analyzing new crimes, and extending its application nationwide.
The platform for the Public Prosecutor's Office will focus on extracting names of people or places, as well as identifying the modus operandi of crimes, through victims' accounts, using social network analysis and AI, to improve criminal prosecution processes.
SOSAFE's data-driven platform will be aimed at police and municipalities and will use predictive models of robberies in different areas to deter crime and design preventive patrol strategies. It will also automate the prioritization of citizen reports for a faster response to urgent situations.