Air Quality Index Prediction Using DNN-Markov Modeling

Abstract

Air quality measurements contribute to diverse socio-economic sectors, including the environment and healthcare. Many methods are commonly applied to present air-quality levels, reflecting differing national standards. This study presents an air quality index prediction model, to measure air pollution levels for healthcare applications in congested areas. DNN-Markov modeling techniques are used to predict air quality, based on environmental conditions at peak hours. The developed model presents different approaches for highly accurate prediction of the air quality index for the next hour at a given location, under specific environmental conditions. This system could be used to support planning decisions related to the consequences of air quality. The study was conducted in selected locations in Jordan and England as a comparative model prediction accuracy study using different big-data sets of multivariate time series in traffic-heavy locations. The air quality index was represented using Neuro Fuzzy Logic as a method to contribute in air quality index predictions within blurry (boundary) values. The selected DNN-Markov hybrid model could predict air quality with accuracy of around (RMSE 7.86) for the location in England, and around (RMSE 15.27) for the one in Jordan.