Automatic Crime Prediction Using Events Extracted from Twitter Posts

You are here

This 2012 article focuses on predictive analytics and in particular crime prediction using events and data extracted from Twitter posts. The authors present a preliminary investigation of Twitter-based criminal incident prediction. Their approach in based on the automatic semantic analysis and understanding of natural language Twitter posts, combined with dimensionality reduction via latent Dirichlet allocation and prediction linear modeling. The authors argue that many current models for predicting crime do not take into account the rich and rapidly expanding social media context that surrounds incidents of interest.

This article will be of interest to PVE researchers and practitioners interested in predictive analytics and modeling. In their study, the authors tested their model on the task of predicting future hit-and-run crimes. The study’s evaluation results indicate that the model outperforms a baseline model that predicts hit-and-run incidents uniformly across all days.

Xiaofeng Wang, Matthew S. Gerber, and Donald E. Brown

2012