Accurate and real-time traffic state forecasting has become an important and challenging task. The correlations of temporal and spatial are important to the accuracy of traffic forecasting. We propose a novel deep learning framework, Spatio-temporal CapsNet-LSTM Neural Network(SCLNN) ,to capture the spatial and temporal dependence. The proposed SCLNN is combined with Capsule network and LSTM. Capsule network is used to capture the complex spatial dependence and the LSTM is used to capture the traffic data temporal dependence. Experiments show that our SCLNN network can capture the spatio-temporal correlations throughout traffic data and the outperform state-of-art baselines algorithms on real-world traffic datasets

CICTP