MSTDFGRN: A Multi-view Spatio-Temporal Dynamic Fusion Graph Recurrent Network for traffic flow prediction

Computers & Electrical Engineering, Год журнала: 2025, Номер 123, С. 110046 - 110046

Опубликована: Янв. 4, 2025

Язык: Английский

---

A Short-Term Traffic Flow Prediction Method Based on Personalized Lightweight Federated Learning

Sensors, Год журнала: 2025, Номер 25(3), С. 967 - 967

Опубликована: Фев. 6, 2025

Traffic flow prediction can guide the rational layout of land use. Accurate traffic provide an important basis for urban expansion planning. This paper introduces a personalized lightweight federated learning framework (PLFL) prediction. has been improved and enhanced to better accommodate data. It is capable collaboratively training unified global model without compromising privacy individual datasets. Specifically, spatiotemporal fusion graph convolutional network (MGTGCN) established as initial learning. Subsequently, shared parameter mechanism employed training. Customized weights are allocated each client based on their data features enhance personalization during this process. In order improve communication efficiency learning, dynamic pruning (DMP) introduced side reduce number parameters that need be communicated. Finally, PLFL proposed in experimentally validated using LPR from Changsha city. The results demonstrate still achieve favorable outcomes even when certain clients lack Moreover, under while preserving distinct characteristics client, significant interference other clients.

Язык: Английский

---

Unbalanced Traffic Accident Severity Prediction: Fusing Vae-Attention and Gcn Algorithm

Опубликована: Янв. 1, 2025

Язык: Английский

---

A comprehensive survey of Federated Intrusion Detection Systems: Techniques, challenges and solutions

Computer Science Review, Год журнала: 2024, Номер 56, С. 100717 - 100717

Опубликована: Дек. 20, 2024

Язык: Английский

---