… used an LSTM model with dynamically optimal time lag to predict short-term traffic (15). To
… The dynamic time lag was then used to derive the lowest mean absolute percentage error (…

… , most prediction methods with current time-series data lead to inefficient predictions when
… In order to deal with this problem, a dynamic multi-interval traffic volume prediction model, …

… dependency among traffic volumes, in DST-GCNN, we use an undirected graph G = (V, A)
to represent the traffic volumes, where the vertex set V represents traffic volumes at different …

… the ANN has been a widely used method in traffic prediction [23, 24]. It has been … traffic
volume, traffic speed, and travel time prediction [25, 26]. Therefore, ANN is used to fit the dynamic …

… traffic flows. In [1], the authors proposed a deep learning novel residual model for the prediction
… In this research work [3], they proposed a novel architecture to estimate traffic volume at …

… Moreover, traffic volume in a road network also affects the influence of neighboring nodes.
For instance, neighboring nodes may not have strong influence on the targeted node in a …

… -efficient dynamic computing resource allocation in CRAN by predicting the wireless traffic
of … In order to model the temporal correlation in this dataset, the average traffic volume ratio (…

… both spatial dependencies and temporal dynamics of traffic flow in deep learning models,
and thereby enables effective learning and accurate speed prediction for network-scale traffic. …

… at modelling and predicting the traffic flow in en route airspace with full consideration of the
dynamics and uncertainty of the air traffic, this paper proposes a novel dynamic air traffic flow …

… (10) pointed out that, depending on the traffic volume level and flow characteristics, the
observed flow within (typically short) time intervals would exhibit different degrees of variability. In …