ABSTRACT
The coronavirus disease already cost the world more than 6.3 million cumulative deaths and approximately 650 million reported cases of infection. The fourth wave is predicted by several organizations, while many nations are already facing the critical conditions of the fourth wave currently. Predicting the pace of COVID-19 and estimating its transmission probability inside closed spaces is crucial for all the stakeholders, policymakers, and health professionals. Based on the relevance of machine learning techniques in estimating the COVID-19 transmission probability, we collected the data through real-time measurements of eleven input parameters. These inputs are used to forecast the R-Event value as a target. Algorithms based on machine learning are employed to forecast the new COVID-19 cases inside an office room. The performances of the six approaches are compared against each other considering conventional statistical indicators. Researchers, policymakers, designers, and other stakeholders can use this study to forecast indoor viral transmission in naturally ventilated office rooms.
