Context- and situation prediction for outdoor air quality monitoring

Abstract

The staggering increase in deaths caused by the rise of air pollution in urban areas is a growing global concern, hence predicting the time and place where concentrations of pollutants will be the highest is critical for air quality monitoring systems. We provide a thorough review of the latest air quality prediction algorithms and show that they are usually focused mainly on improving the forecasting algorithms themselves, leaving valuable contextual information aside. Thus, we introduce a context-aware computing model for outdoor air quality monitoring and prediction systems. We design and describe a novel context and situation reasoning model, that considers external environmental context, specifically traffic volumes and fire incidents, along with user based context attributes, to feed into a state-of-the-art machine learning prediction model. We demonstrate the adaptability and customisability of the proposed design in the implementation of our responsive My Air Quality Index (MyAQI) web application, that shifts the focus towards the individual needs of each end-user, without neglecting the benefits of the latest air pollution forecasting algorithms. We test the implementation with different user profiles and show the results of the system’s adaptation. We also demonstrate the prediction model accuracy, when considering user and extended environmental context, for 4 air quality monitoring stations in the Melbourne Region in Victoria, Australia.