Application of selected AI models to reconstruction of noise levels at thoroughfare during selected year

The work briefly presents 3 selected models of noise level at thoroughfare. The models were built by using equivalent sound level data recorded by monitoring station. The methods applied for building the models were Random Forest and regression trees. Next, the obtained models were used for reconstruction of equivalent sound levels for over 60 days for which monitoring station did not record the sound level. Noise level values were reconstructed separately for the following 24-hour sub-intervals: day (6-18), evening (18-22) and night (22-6). The statistical analysis of the sound levels during the whole year, before and after reconstruction, was carried out.