A Modified γ-Sutte Indicator for Air Quality Index Prediction

Air pollution has become an essential issue in environmental protection. The Quality Index (AQI) is often used to determine the severity of air pollution. When AQI reaches red level, proportion asthma patients seeking medical treatment will increase by 30% more than usual. If can be predicted advance, benefits early warning achieved. In recent years, a scholar proposed α-Sutte indicator which shows its excellence time series prediction. However, calculation indicators uses fixed weight. Thus, β-Sutte indicator, using dynamic weight with high computation cost, appeared. computational complexity and sliding window required are still compared indicator. this study, modified γ-Sutte lower cost proposed. order prove that good generalization ability transferable, study data from different regions periods predict AQI. results showed prediction accuracy was better other methods.