Latent Factorization of Tensors with Threshold Distance Weighted Loss for Traffic Data Estimation
This work addresses data quality issues in intelligent transportation systems, but it is incremental as it modifies an existing method to handle outliers.
The paper tackled the problem of missing or corrupted spatiotemporal traffic data in intelligent transportation systems by proposing a TDW loss-incorporated latent factorization of tensors model, which consistently outperformed state-of-the-art approaches in prediction accuracy and computational efficiency on two traffic speed datasets.
Intelligent transportation systems (ITS) rely heavily on complete and high-quality spatiotemporal traffic data to achieve optimal performance. Nevertheless, in real-word traffic data collection processes, issues such as communication failures and sensor malfunctions often lead to incomplete or corrupted datasets, thereby posing significant challenges to the advancement of ITS. Among various methods for imputing missing spatiotemporal traffic data, the latent factorization of tensors (LFT) model has emerged as a widely adopted and effective solution. However, conventional LFT models typically employ the standard L2-norm in their learning objective, which makes them vulnerable to the influence of outliers. To overcome this limitation, this paper proposes a threshold distance weighted (TDW) loss-incorporated Latent Factorization of Tensors (TDWLFT) model. The proposed loss function effectively reduces the model's sensitivity to outliers by assigning differentiated weights to individual samples. Extensive experiments conducted on two traffic speed datasets sourced from diverse urban environments confirm that the proposed TDWLFT model consistently outperforms state-of-the-art approaches in terms of both in both prediction accuracy and computational efficiency.