Assessment of risks of pandemics to communities and workplaces requires an intelligent decision support system (DSS). The core of such DSS must be based on machine reasoning techniques such as inference and shall be capable of estimating risks and biases in decision making. In this paper, we use a causal network to make Bayesian inference on COVID-19 data, in particular, assess risks such as infection rate and other precaution indicators. Unlike other statistical models, a Bayesian causal network combines various sources of data through joint distribution, and better reflects the uncertainty of the available data. We provide an example using the case of the COVID-19 outbreak that happened on board of USS Theodore Roosevelt in early 2020.
- Pub Date:
- August 2020
- Computer Science - Computers and Society
- 9 pages, 3 figures, submitted to IEEE Transactions on Emerging Topics in Computational Intelligence. arXiv admin note: text overlap with arXiv:2008.03845