Crime prediction for patrol routes generation using machine learning

Citizen security is one of the main objectives of any government worldwide. Security entities make multiple efforts to apply the latest technologies in order to prevent any type of criminal offence. The analysis of a database of the National Police of Ecuador has allowed us generating patrol routes to prevent and reduce the crime rate in the city of Quito, Ecuador. The K-means clustering has been used to determine the points of greatest crime concentration and then linear regression is applied for the prediction of crimes within subgroups of data. Those way-points will allow to generate and optimize police patrol routes. The results obtained in the prediction of crimes is greater than 80%. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.