Mobile cell traffic forecast

The brief

Network planners add capacity after customers complain. Forecast internet traffic for each area of a city with daily and weekly patterns, find the areas heading for congestion, and recommend where to add capacity. Good looks like: forecasts that beat a seasonal baseline across many cells, back-testing, and a congestion watch list.

What you will build

1. Prepare traffic series per grid cell
2. Explore daily and weekly patterns
3. Compare seasonal-naive and gradient-boosted forecasts
4. Back-test across many cells
5. Flag cells heading for congestion

Free, open data to build it with

• Telecom Italia big data challenge (Milan telecommunication activity) ↗ Harvard Dataverse

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.