Mukara: Highway Traffic Volume Data for England and Wales

Eight years of data from England and Wales used to train a deep learning model for traffic volume prediction. The dataset incorporates population, employment, land use, road network characteristics, and points of interest as inputs. It was authored by Yue Li and last updated in April 2026.

Use Cases

• Training deep learning models for traffic volume prediction based on socioeconomic and network features
• Evaluating spatial transferability of prediction models using region-based cross-validation
• Analyzing the relative importance of different input feature groups (e.g., population, land use) for traffic forecasting
• Benchmarking new traffic prediction frameworks against established models and studies

Strengths

• Model trained on eight years of data, providing a substantial temporal scope
• Incorporates multiple input feature groups: population, employment, land use, road network, and points of interest
• Performance metrics reported: mean GEH of 50.74, mean absolute error of 8,989 vehicles/day, and R² of 0.583

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Data may reflect geographic bias inherent to the England and Wales source region

Provenance

Source

figshare

Collection Method

Data used to train the proposed Mukara deep learning framework.

Time Range

Eight years (specific years not stated)

Freshness

Last updated 2026-04-16 17:41:56; freshness should be verified

Geography

England and Wales