Mini-project
Energy, Power & Utilities
Electricity theft detection
Find the few meters with likely power theft among thousands, and explain each case.
The brief
Distribution companies lose a large share of power to theft. Using daily consumption records with known theft labels, find the meters most likely to be stealing, handle the heavy imbalance, and explain each case so inspectors visit the right homes. Good looks like: metrics for an imbalanced problem, a ranked inspection list, and explanations inspectors understand.
What you will build
- Explore consumption patterns of honest and theft customers
- Build features from daily usage curves
- Train gradient boosting with imbalance handling
- Evaluate precision at the inspection team's capacity
- Explain cases with SHAP
Free, open data to build it with
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.