An Event-Centric Framework for Predicting Crime Hotspots With Flexible Time Intervals
IEEE Transactions on Knowledge and Data Engineering · 2025-12 · 37 (12) · 6902-6915
DOI: 10.1109/tkde.2025.3618389 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Knowledge and Data Engineering, 2025-12. Verfasst von Jiahui Jin und 5 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Major Special Program of Jiangsu, Jiangsu Provincial Key Laboratory of Network and Information Security.
Abstract (Originalsprache)
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 ${\sf FlexiCrime}$ FlexiCrime , a novel event-centric framework for predicting crime hotspots with flexible time intervals. ${\sf FlexiCrime}$ 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. Together, the crime context and evolving features allow us to predict whether an urban area is a crime hotspot given a future time interval. To evaluate ${\sf FlexiCrime}$ 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.
Förderung
- National Natural Science Foundation of China
- Science and Technology Major Special Program of Jiangsu
- Jiangsu Provincial Key Laboratory of Network and Information Security
- Key Laboratory of Computer Network and Information Integration of the Ministry of Education of China
- Collaborative Innovation Center of Novel Software Technology and Industrialization
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Jin, J., Hong, Y., Xu, G., Zhang, J., Tang, J., & Wang, H. (2025). An Event-Centric Framework for Predicting Crime Hotspots With Flexible Time Intervals. IEEE Transactions on Knowledge and Data Engineering, 37(12), 6902-6915. https://doi.org/10.1109/tkde.2025.3618389
Vancouver
Jin J, Hong Y, Xu G, Zhang J, Tang J, Wang H. An Event-Centric Framework for Predicting Crime Hotspots With Flexible Time Intervals. IEEE Transactions on Knowledge and Data Engineering. 2025;37(12):6902-6915. doi:10.1109/tkde.2025.3618389
BibTeX
@article{Jin2025,
author = {Jin, Jiahui and Hong, Yi and Xu, Guandong and Zhang, Jinghui and Tang, Jun and Wang, Hancheng},
title = {An Event-Centric Framework for Predicting Crime Hotspots With Flexible Time Intervals},
journal = {IEEE Transactions on Knowledge and Data Engineering},
year = {2025},
volume = {37},
number = {12},
pages = {6902--6915},
doi = {10.1109/tkde.2025.3618389}
}
EUVIMED