Crime Rate Inference with Big Data

• DOI: 10.1145/2939672.2939736
• Publication Date: 2016-08-08T18:33:46Z
• Authors: Hongjian Wang; Daniel Kifer; Corina Graif; Zhenhui Li
• Keywords: Demographics; Crime rate

Crime is one of the most important social problems in country, affecting public safety, children development, and adult socioeconomic status. Understanding what factors cause higher crime critical for policy makers their efforts to reduce increase citizens' life quality. We tackle a fundamental problem our paper: rate inference at neighborhood level. Traditional approaches have used demographics geographical influences estimate rates region. With fast development positioning technology prevalence mobile devices, large amount modern urban data been collected such big can provide new perspectives understanding crime. In this paper, we large-scale Point-Of-Interest taxi flow city Chicago, IL USA. observed significantly improved performance compared using traditional features. Such an improvement consistent over multiple years. also show that these features are significant feature importance analysis.