Forecasting of the daily meteorological pollution using wavelets and support vector machine

摘要: The paper presents the method of daily air pollution forecasting by using support vector machine (SVM) and wavelet decomposition. Based on observed data NO"2, CO, SO"2 dust, for past years actual meteorological parameters, like wind, temperature, humidity pressure, we propose approach, applying neural network SVM type, working in regression mode. To obtain acceptable accuracy forecast decompose measured time series into representation predict coefficients. On basis these predicted values final is prepared. results numerical experiments measurements made stations, situated northern region Poland.