STGNN-FAM: A Traffic Flow Prediction Model for Spatiotemporal Graph Networks Based on Fusion of Attention Mechanisms

Figure 16

Visualization of the prediction results of node 1 after introducing missing values. (a) Forecast results of node 1 when the missing rate is 1%. (b) Forecast results of node 1 when the missing rate is 2%. (c) Forecast results of node 1 when the missing rate is 3%. (d) Forecast results of node 1 when the missing rate is 4%. (e) Forecast results of node 1 when the missing rate is 5%.

(a)

(b)

(c)

(d)

(e)