|Home||<< 1 >>|
Binxu Zhai, & Jianguo Chen. (2017). Research on the forecasting of Air Quality Index (AQI) based on FS-GA-BPNN: A case study of Beijing, China. In eds Aurélie Montarnal Matthieu Lauras Chihab Hanachi F. B. Tina Comes (Ed.), Proceedings of the 14th International Conference on Information Systems for Crisis Response And Management (pp. 307–321). Albi, France: Iscram.
Abstract: The analysis and forecasting of eminent air quality play a significant role in municipal regulatory planning and emergency preparedness. In this paper, a FS-GA-BPNN model forecasting the daily average Air Quality Index (AQI) is proposed. Special procedures for feature extraction to find more potential significant variables and feature selection to remove redundant information and avoid overfitting are conducted before modelling. Three different models – BPNN, GA-BPNN and FS-GA-BPNN are established to compare the prediction accuracy, generalization ability and reliability. 17 parameters involving pollutant concentration, meteorological elements and surrounding factors are found essential for the method effectiveness. The result shows that the FS-GA-BPNN model generally performs superior to ordinary BPNN, suggesting the necessity of extensive data mining and feature extraction for successful machine learning. The results of this paper can help to conduct air quality pre-warning system and improve the emergency planning process of extreme weather events.