Linking of open and private data in dataspace: A case study of air quality monitoring and forecasting

Air pollutant monitoring and its efficient visualisation can support accurately assessing the air quality and harmful emissions; and it can guide us towards potential mitigation strategies to reduce its impact on public health and our environment. This paper presents a case study of employing dataspaces and proposing an ontology for modelling mobility to address the challenges posed by the heterogeneity of data sources in environmental monitoring, as well as using machine learning for forecasting pollutants. We employ Linked data as a powerful paradigm for harmonising and interlinking diverse and publicly available environmental data with private company data to create a dataspace for environmental monitoring. By applying semantic technologies and ontological modelling to integrate heterogeneous data, our approach fosters data interoperability and facilitates enhanced data exploration and decision support. For decision support, we demonstrate the utility of integrated data for forecasting air pollutants with the help of models developed using machine learning. Finally, a spatio-temporal visualisation platform harnesses the power of semantic relationships and contextual enrichment to support data exploration.