PeMSD8: Traffic Flow Prediction Data for Intelligent Transportation Systems

PeMSD8 is a traffic flow prediction dataset used to benchmark models like the STF-GGRU, which achieved an RMSE of 11.1 on it. The dataset is hosted on figshare under a CC-BY-4.0 license and was last updated on 2026-04-28. Author Karimeh Ibrahim Ata contributed this 5.5 KB Excel file.

Use Cases

• Benchmarking traffic prediction models based on the reported RMSE performance metrics.
• Training graph neural networks to capture non-Euclidean road network relationships mentioned in the description.
• Developing real-time adaptive systems for Intelligent Transportation Systems based on dynamic traffic patterns.
• Studying spatiotemporal feature alignment techniques using integrated modules like ISTFA described in the research.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and initial exploration.
• Serves as a benchmark for advanced models, with a documented STF-GGRU RMSE score of 11.1.
• Available under a permissive CC-BY-4.0 license for broad reuse.

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment.
• The description focuses on a research model's performance; specific data attributes like sensor count or time range are not provided.
• Data freshness should be verified despite the 2026 update date, as the underlying collection period is unspecified.

Provenance

Source

figshare, contributed by Karimeh Ibrahim Ata.

Collection Method

Likely sourced from the PeMS (Performance Measurement System) traffic data collection network.

Time Range

null

Freshness

Last updated 2026-04-28 17:37:43; freshness should be verified.

Geography

null