CODS 624 Space-Time Data Science
Space-time data are becoming available in overwhelming volumes and diverse forms as a result of growing remote-sensing capabilities, ground-based sensor networks, crowdsourcing, citizen science data, climate models, and novel medical sensing technologies. Dealing with massive data sets having complex structures implies a collection of conceptual, methodological, and technical challenges, which are exacerbated by the data diversity. Space-time statistical methods were not designed to deal with global, high-volume, hyper-dimensional, heterogeneous and uncertain space-time data. In fact, the computational requirements of most available methods scale poorly with data size. Space-Time Data Science (STDS throughout) is based on the integration of Statistics, Computer Science and Machine Learning as fundamental vertices in a graph structure to be then synchronized with applied sciences, such as geography, physics, soil science, neuroscience, and epidemiology. Hence, the key of success of STDS is to be able to tailor interdisciplinary approaches to the analysis of diverse and big space-time data. This course will introduce the statistical and computational aspects of STDS.