How to Identify Patterns of Citywide Dynamic Traffic at a Low Cost? An In-Depth Neural Network Approach with Digital Maps

<div>Distance change during peak times: take the first vector in morning and evening peak time as the baseline. The horizontal axis represents the number of samples (a time series). The vertical axis represents the distance between vectors. (a) Distance evolution during morning peak times and (b) distance evolution during evening peak times.</div>

Complexity

fig11

Figure 11

Figure 11: How to Identify Patterns of Citywide Dynamic Traffic at a Low Cost? An In-Depth Neural Network Approach with Digital Maps 