NorwayPublic sector2020 – ongoingNorwegian Tax Administration
Machine learning for undeclared rental-income checks and guidance
Norway's tax administration attributed over NOK 110 million in additional declared income to a rental model and follow-up measures
Problem
Some owners of secondary homes did not report taxable rental income. The administration needed to direct checks and guidance toward likely omissions.
Approach
The administration developed an XGBoost model for tax year 2020 using information about owners and properties. It ranked owners by rental probability to support audit selection. The model also helped select recipients of information letters.
Results
Takeaway
The reported income increase combined model-assisted selection with follow-up measures. It represented additional declared income rather than tax collected.
Categories
- Industry
- Public sector
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
The administration described its rental-probability model and its use in checks and guidance. It attributed over NOK 110 million in additional declared ordinary income to the model and related measures.
Full details are in the original sources.
