Impacts of Environmental Factors on Wellbeing: Machine Learning-Based Benchmarking of Spatial and Temporal Properties

Environmental noise impacts wellbeing with consequences on physical and mental health. Chronic exposure to noise pollution can amplify stress levels, disturb sleep patterns, and increase the risk of cardiovascular issues. Individuals subjected to continuous noise often report high irritability, impaired concentration, and diminished cognitive performance which may lead to anxiety and depression, particularly in vulnerable populations, such as children and the elderly. Recognising the profound impact of environmental noise is crucial for public health, necessitating comprehensive strategies to mitigate and regulate noise pollution for the enhancement of overall wellbeing. Despite the recent advancement of artificial intelligence (AI), this issue is under-addressed as no benchmark dataset is available to perform an in-depth analysis of the factors that subtly and gradually contribute to the deterioration of physical and mental wellbeing. To fill this gap, the current work presents a benchmarking dataset and explores the spatial and temporal properties of the same in understanding how these environmental factors affect wellbeing. Experiments with explainable AI (XAI) indicate that the Random Forest and Gradient Boosting classifiers can be considered as baseline XAI methods to analyse the outdoor and indoor environmental data, respectively.