A Hybrid Deep Learning Model for Short-Term Traffic Flow Pre-Diction Considering Spatiotemporal Features

Traffic flow prediction is one of the basic, key problems with developing an intelligent transportation system since accurate and timely traffic can provide information support decision for control guidance. However, due to complex characteristics information, it still a challenging task. This paper proposes novel hybrid deep learning model short-term by considering inherent features data. The proposed consists three components: recent, daily weekly components. recent component integrated improved graph convolutional network (GCN) bi-directional LSTM (Bi-LSTM). It designed capture spatiotemporal features. remaining two components are built multi-layer Bi-LSTM. They developed extract periodic focus on important using attention mechanism. We tested performance our real-world dataset experimental results indicate that has better than those previously.