Traffic Flow Data for Kandovan Mountain Road with Weather and Event Features, 2017-2023

Eight machine learning models, including Deep LSTM and Random Forest Regressor, were trained on traffic, weather, and event data from 2017 to 2023 to predict flow on Kandovan Road. The Random Forest Regressor achieved the highest accuracy with an R² score of 0.88. The work resulted in a traffic forecasting software system for real-time visualization and decision support.

Use Cases

• Benchmarking machine learning models for traffic prediction based on the eight implemented algorithms mentioned in the description
• Analyzing the impact of weather conditions on traffic flow in mountainous suburban areas
• Developing decision-support tools for traffic authorities based on the described forecasting software system
• Studying non-linear and dynamic traffic patterns influenced by calendar events and road-specific characteristics

Strengths

• Integrates traffic, weather, and event data from 2017 to 2023, a 6-year time range
• Model performance is quantified with specific metrics including an R² score of 0.88 for the best model
• Focuses on a specific, challenging geography (Kandovan Road, a mountainous suburban route)

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 13.5 KB, indicating limited scope or a summary of results rather than raw observational data

Provenance

Source

figshare, authored by Shahriar Afandizadeh

Collection Method

Likely collected from traffic sensors and weather sources for the specified road.

Time Range

2017 to 2023

Freshness

Last updated 2026-05-05 17:24:57; freshness should be verified

Geography

Kandovan Road, a mountainous suburban route