Predictive Analytics for Supporting Environmental Sustainability and Disaster Management
Adam B. Bouttell, Sijin Lee, Carson Kai-Sang Leung, Martin J. Levesque, Seunggon Son, Weihong Zhang, Alfredo Cuzzocrea · 2022 IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC) · 2022
As we are living in an uncertain world, the uncertainty may have significant and/or direct implications on various aspects of the computer industry. Hence, innovations of computers, software and applications have emerged as a pressing need. Data science solutions have been designed and/or developed for social good. For example, in this paper, we present a predictive analytics solution for supporting environmental sustainability. In particular, we focus on environmental sustainability and disaster management related to water levels. To elaborate, low water levels may lead to unsustainability in water supply. In contrast, high water levels may lead to hazards or disasters like floods. Thus, having a reliable predictive analytics solution that gives accurate estimates of water levels is important. Our solution integrates different categories of weather data collected from distributed rich data sources. Evaluation on real-life data from a Canadian city demonstrates the practicality of our solution in predicting water level, and thus supports environmental sustainability and disaster management.