Abstract:
Internet of things sensors are enabling data-driven traffic speed prediction, a foundation of advanced traffic management. However, current research studies mostly predict traffic for one hour ahead. Long-term prediction methods have error accumulation, exposure bias, or low-granularity future data. Based on recent graph deep learning techniques, this paper proposes a data-driven, long-term, high-granularity traffic speed prediction approach. A predictor-regularizer architecture embeds traffic dynamics spatial-temporal data correlation into the prediction process. Both subnetworks use graph convolutions for geometrical latent information extraction and reconstruction. Comprehensive case studies on real-world datasets show that the proposed approach consistently outperforms baselines. It pioneered network-wide long-term traffic speed prediction. The proposed approach’s design principles can inform deep learning-based transportation research.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
