Fine-Grained Crime Prediction in an Urban Neighborhood


Crime is a serious problem that has severe implications on city resources. In this work, we present a novel probabilistic model that predicts the occurrence of various crime types by learning from previous crime incidents and takes advantage of joint dependencies across crime types. We perform a preliminary evaluation using a real-world dataset of crime incidents reported in Memphis across several precincts which shows the promise of our approach.

Publication Title

2018 IEEE International Smart Cities Conference, ISC2 2018