Abstract
Covid-19 is a global disaster which requires not only medical humanity but also computing power to predict, and, ideally, interpret these pandemics. Deep models can be trained to be fairly accurate. However, the mechanism of models prevents from explainability. On the hand, epidemiological approaches, e.g. SIR, help experts with insightful information. However, those need to be provided with parameter values, which become complicated in certain locations. The fourth wave of the pandemic in Ho Chi Minh City (HCMC), Vietnam in 2021 offers valuable lessons along with real and specific data. Hence, we introduce an explainable AI model known as BeCaked+ to predict and analyze efficiently the pandemic situation from the collected data. BeCaked+ combined deep learning and epidemiological models enhanced by specific parameters related to the policies endorsed by the government. Such combination makes BeCaked+ accurate and informative for policymakers to make appropriate responses. One take a try BeCaked+ at http://www.cse.hcmut.edu.vn/BeCaked.
Duc Q. Nguyen
CS Master Student
My research interests include Generative Models, Graph Representation Learning, and Probabilistic Machine Learning. My application interests include Natural Language Processing, Healthcare, and Education.