<p>Studying the temporal change characteristics and influencing factors of air pollutants and implementing collaborative treatment is of critical importance for the improvement of air quality. This study examines the dynamic characteristics of particulate matter (PM_2.5, PM₁₀) and gaseous pollutants (O₃, NO₂, CO, SO₂) and their influencing factors in Jinan and Qingdao from 2014 to 2023 are investigated based on KZ filtering and random forest regression (RFR) models. The results show that the long-term components of PM_2.5, PM₁₀, SO₂, NO₂ and CO all show a significant decreasing trend, while O₃ shows a decreasing trend from 2014 to 2016 and a fluctuating increasing trend after 2016. The R² of RFR between short-term air pollutants and meteorological factors in Jinan and Qingdao ranged from 0.572 to 0.874, and for baseline components, from 0.432 to 0.926 and 0.435 to 0.881, respectively. Significantly superior accuracy was observed in the RFR model relative to the stepwise regression model. Meteorological factors significantly affected the short-term and baseline components of air pollutants, and there were regional differences in the degree of influence. There is a significant correlation between meteorological factors and air pollutant concentrations, with atmospheric temperature having the most significant effect on the baseline component of air pollutant concentrations. Air pollutant concentrations in the two cities are mainly affected by anthropogenic emissions, and the contribution of anthropogenic emissions to air pollutant concentrations is 75% and 80%, respectively, but has been decreasing in recent years, while the influence of meteorological factors has been on the rise.</p>