University Gym Crowdedness with Weather and Semester Data

A university campus gym recorded 26,000 people counts at 10-minute intervals over one year. The data includes timestamp, day of week, weather, and semester information to predict occupancy. It was collected with consent from the university and the gym.

Use Cases

• Predict gym occupancy based on time of day and weather features mentioned in the description.
• Identify key factors affecting crowdedness based on semester start and holiday flags.
• Optimize personal workout schedules by forecasting low-occupancy periods.
• Analyze temporal patterns in student behavior linked to academic calendars.

Strengths

• Contains 26,000 observations providing a substantial sample for analysis.
• Includes multiple predictive features such as weather, time, and academic calendar indicators.
• Data collection was conducted with explicit consent from the involved institutions.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Last update date is unknown; freshness unverified.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

openml

Collection Method

People counts measured every 10 minutes, with extra info gathered on weather and semester specifics.

Time Range

Over the last year (specific dates unknown).

Geography

A university campus gym (specific location unknown).