Learning A Universal Crime Predictor with Knowledge-guided Hypernetworks

TL;DR

This paper introduces HYSTL, a hypernetwork-based framework that leverages crime knowledge graphs to improve urban crime prediction across diverse cities with varying crime data, outperforming existing methods.

Contribution

The paper presents a novel hypernetwork approach guided by crime knowledge graphs to unify crime prediction across cities with different crime types.

Findings

HYSTL outperforms state-of-the-art baselines in experiments.

The framework effectively handles non-overlapping crime types across cities.

Knowledge-guided hypernetworks enhance prediction accuracy.

Abstract

Predicting crimes in urban environments is crucial for public safety, yet existing prediction methods often struggle to align the knowledge across diverse cities that vary dramatically in data availability of specific crime types. We propose HYpernetwork-enhanced Spatial Temporal Learning (HYSTL), a framework that can effectively train a unified, stronger crime predictor without assuming identical crime types in different cities' records. In HYSTL, instead of parameterising a dedicated predictor per crime type, a hypernetwork is designed to dynamically generate parameters for the prediction function conditioned on the crime type of interest. To bridge the semantic gap between different crime types, a structured crime knowledge graph is built, where the learned representations of crimes are used as the input to the hypernetwork to facilitate parameter generation. As such, when making predictions for each crime type, the predictor is additionally guided by its intricate association with other relevant crime types. Extensive experiments are performed on two cities with non-overlapping crime types, and the results demonstrate HYSTL outperforms state-of-the-art baselines.