| Abstract |
Classification models to forecast exceedance of the ozone (O-3) threshold established by European legislation are rare in literature, as is the focus on background O-3, with higher concentrations at city outskirts. This study evaluated the performance of nine classifiers to forecast this threshold exceedance by background O-3. Models used five large hourly background O-3 data sets (2006-2015), and included temporal features describing the O-3 formation dynamic. Bagging and stacking ensembles of such classifiers and their cost of learning were also evaluated. C5.0 and nnet classifiers achieved the best forecasting performance, even at imbalanced learning. Bagging ensembles outperformed stacking approaches, although with little accuracy improvement as compared to classifiers. The cost of learning evidenced similar performance results from reduced fractions of original data sets. The use of these models to forecast background O-3 threshold exceedances are encouraged due to the performances obtained and to their easy reproducibility. |
