An Event-centric Framework for Predicting Crime Hotspots with Flexible Time Intervals

TL;DR

FlexiCrime is a novel event-centric framework that predicts urban crime hotspots over flexible time intervals by capturing complex spatiotemporal correlations and crime type evolutions, outperforming existing fixed-interval methods.

Contribution

Introduces FlexiCrime, a new framework combining continuous-time attention and type-aware point processes for flexible crime hotspot prediction.

Findings

Outperforms baseline models in real-world datasets.

Effectively captures crime patterns across different time intervals.

Accurately predicts crime hotspots for multiple crime types.

Abstract

Predicting crime hotspots in a city is a complex and critical task with significant societal implications. Numerous spatiotemporal correlations and irregularities pose substantial challenges to this endeavor. Existing methods commonly employ fixed-time granularities and sequence prediction models. However, determining appropriate time granularities is difficult, leading to inaccurate predictions for specific time windows. For example, users might ask: What are the crime hotspots during 12:00-20:00? To address this issue, we introduce FlexiCrime, a novel event-centric framework for predicting crime hotspots with flexible time intervals. FlexiCrime incorporates a continuous-time attention network to capture correlations between crime events, which learns crime context features, representing general crime patterns across time points and locations. Furthermore, we introduce a type-aware spatiotemporal point process that learns crime-evolving features, measuring the risk of specific crime types at a given time and location by considering the frequency of past crime events. The crime context and evolving features together allow us to predict whether an urban area is a crime hotspot given a future time interval. To evaluate FlexiCrime's effectiveness, we conducted experiments using real-world datasets from two cities, covering twelve crime types. The results show that our model outperforms baseline techniques in predicting crime hotspots over flexible time intervals.