LSTM Networks in Traffic Forecasting and Optimization: Capabilities, Challenges and Future Prospects

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Recent advancements in LSTM technology improve traffic flow prediction and management, highlighting challenges and future potential.

LSTM-based models achieve over 90% accuracy in urban traffic forecasting, improving predictive performance significantly.
Hybrid models like GCN-LSTM exhibit a 95.2% fitting degree, reducing prediction errors by 30% compared to standalone models.
Innovations such as attention mechanisms reduce mean absolute error to 0.401 in traffic datasets, enhancing prediction robustness.
Real-time LSTM frameworks have shown applications in air traffic safety, reducing flight conflict prediction errors to less than 3%.

SMETIUKH et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark