Deep Learning for Urban Traffic Prediction

Abstract

We present a novel deep learning architecture for predicting urban traffic patterns using multi-modal sensor data. Our approach combines convolutional neural networks with attention mechanisms to capture both spatial and temporal dependencies in traffic flow. Experiments on real-world datasets from three major cities demonstrate that our method outperforms existing approaches by 15% in prediction accuracy while reducing computational requirements by 40%.

Citation

@article{smith2024traffic,
  title = {Deep Learning for Urban Traffic Prediction},
  author = {Smith, Jane and Doe, John and Johnson, Alice},
  journal = {IEEE Transactions on Intelligent Transportation Systems},
  year = {2024},
  doi = {10.1109/TITS.2024.0001}
}