CrimeRadar is a digital platform that uses advanced machine learning to predict crime rates in different neighborhoods at different times. It runs on smartphones and desktop browsers. The first version is focused on metropolitan Rio de Janeiro. CrimeRadar draws on over five years of crime data (roughly 14 million reported crime events) collected by the Rio de Janeiro state police. By isolating out patterns across incidents, locations and time, the app is able to determine relative crime risks today, tomorrow and into next week – and also features historical crime trends going back 12 months. The app was conceived and developed by Igarapé Institute, associated with Via Science and Mosaico.


What is CrimeRadar?

CrimeRadar is an application that forecasts the probability of crime. It runs on smartphones and desktop browsers. The software uses advanced data analytics to show relative crime rates for different neighborhoods at different times in the City of Rio de Janeiro. The data is displayed in an interactive heat map. The app aims to empower the public, engaging the public in crime prevention.

Why use CrimeRadar?

CrimeRadar was developed to improve access to information and promote transparency in its use by locals and visitors. A key is also to make such data accessible, interactive and useful. Until recently, basic crime data in Rio de Janeiro was inaccessible to the wider public. The app works in favor of accessibility and interactivity of such data. CrimeRadar provides real-time data to users, calculating safety levels at different locations and times of day for the city.

How CrimeRadar deals with forecasting crimes?

CrimeRadar does not pin-point specific geolocation points of future crime, but rather uses 250m x 250m areas. What is more, it does not include any racial or subject profiling data. Instead, it restricts the analysis to the frequency and seasonality of reported crimes by specific location, date and time. CrimeRadar seeks to provide the public a way to visualize crime data while in transit through different points of the city. It processes GPS coordinates and a specific time and date and returns a score of the relative probability of a crime at that location and time. This score is then summarized on a scale of 1 to 10 – where 1 is low risk and 10 is high risk – in micro-areas of the city. As a rule of thumb, any time one travels somewhere new they should always exercise caution and be aware of their surroundings.


Featured by

CrimeRadar em reportagem do The Guardian