For DSTMIN

Missing traffic data caused by sensor failures or communication errors significantly hinders the efficiency of downstream tasks in Intelligent Transportation Systems (ITS), such as the critical functions of traffic monitoring and decision-making. Considering the complex distribution of missing data, it is essential to incorporate the missing features to extract dynamic spatial-temporal correlations in traffic processes. Motivated by these concerns, a novel Dynamic Spatial-Temporal Imputation Network with Missing Features (DSTMIN) is proposed to accurately impute traffic data. DSTMIN comprises an embedding layer, a Mask Attention module (MA), and a Fusion Graph Convolution module (FGC). Specifically, an embedding layer is designed to accurately represent the distribution of missing data, thereby capturing both temporal features and missing features. Meanwhile, in order to capture the temporal correlations, MA incorporates missing features to accentuate the impact of observed data and mitigate the interference caused by incomplete data. The proposed DSTMIN is evaluated on two datasets to showcase its superior performance. Specifically, compared to state-of-the-art baselines, DSTMIN achieves a remarkable 20% reduction in imputation error.