Machine learning predicts hourly indoor ozone concentrations with high accuracy. (IMAGE)
Caption
Researchers developed a random forest model that predicts hourly indoor ozone (O₃) levels using easily accessible predictors—ambient ozone estimates, meteorological data, and window-opening behavior—derived from low-cost sensor measurements across 18 Chinese cities. The model achieved strong performance (R² = 0.83, RMSE = 7.21 ppb), demonstrating that incorporating ventilation status significantly improves prediction accuracy and provides a practical approach for large-scale indoor air quality assessment.
Credit
Eco-Environment & Health
Usage Restrictions
Credit must be given to the creator. Only noncommercial uses of the work are permitted. No derivatives or adaptations of the work are permitted.
License
CC BY-NC-ND