Procedia Computer Science, Journal Year: 2025, Volume and Issue: 258, P. 893 - 902
Published: Jan. 1, 2025
Language: Английский
Procedia Computer Science, Journal Year: 2025, Volume and Issue: 258, P. 893 - 902
Published: Jan. 1, 2025
Language: Английский
Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 121, P. 236 - 247
Published: March 1, 2025
Language: Английский
Citations
0Frontiers in Physics, Journal Year: 2025, Volume and Issue: 13
Published: April 2, 2025
To address the challenge of efficiently integrating multi-source heterogeneous data to improve accuracy public safety event prediction, this study proposes and validates a novel prediction model, GATPNet, based on data. The model integrates Graph Attention Networks (GAT), Spatiotemporal Transformers, Proximal Policy Optimization (PPO) achieve effective fusion, spatiotemporal feature extraction, real-time decision support. Through experiments conducted Los Angeles Crime Data CrisisLexT26 datasets, demonstrates that GATPNet outperforms other baseline models. On dataset, achieved an 90%, recall 89%, Prediction Accuracy (STPA) 80%, response time 1.9 s, showing 5% improvement in 10% STPA over best method. it 88%, 78%, 2.1 4% 6% Additionally, ablation further indicate each module plays critical role improving overall performance. Despite model’s high computational complexity when handling large-scale limited coverage still its broad application potential management, offering technical support for social governance emergency management.
Language: Английский
Citations
0Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 125, P. 566 - 574
Published: April 23, 2025
Language: Английский
Citations
0Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 120, P. 648 - 656
Published: Feb. 25, 2025
Language: Английский
Citations
0Procedia Computer Science, Journal Year: 2025, Volume and Issue: 258, P. 893 - 902
Published: Jan. 1, 2025
Language: Английский
Citations
0