Barcelona Accident and Metro Interstate Traffic Volume Data

A combined dataset used to evaluate a hybrid deep learning framework for Intelligent Transportation Systems. The data includes traffic volume from the Metro Interstate and accident data from Barcelona, as referenced in the description. The dataset was authored by Mohammed Saad Javeed and last updated on 2026-05-29.

Use Cases

• Evaluating hybrid deep learning models based on spatiotemporal traffic data.
• Benchmarking anomaly discovery algorithms based on accident and traffic volume data.
• Comparing model performance against baselines like Random Forest and CNN-LSTM.
• Conducting ablation studies on the impact of API fusion and GCN modeling.
• Analyzing the trade-off between inference latency and model accuracy for real-time deployment.

Strengths

• The proposed framework achieved an accuracy of 0.882 and an F1-score of 0.864 on the combined dataset.
• For anomaly discovery, it obtained an AUROC of 0.836 on the Barcelona dataset and 0.819 on the Metro dataset.
• The ablation study provides concrete performance degradation metrics, such as a -4.1% accuracy drop when API fusion is removed.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 5.5 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Likely compiled from public datasets (Barcelona Accident dataset, Metro Interstate Traffic Volume dataset) for research experiments.

Freshness

Last updated 2026-05-29 17:31:59; freshness should be verified.

Geography

Barcelona and a Metro Interstate region (specific location not detailed).