An evidence-based culture: COVID-19 positivity factors during the asymptomatic occurrence in Jakarta, lndonesia[Application of Bayesian Logistic Regression to Mining Biomedical Data]

Coronavirus disease 2019 (COVID-19) has been a global disaster, with over 746,312 confirmed cases and still counting in Indonesia, especially Jakarta, which has about 50 per cent asymptomatic confirmed cases. This paper aims to investigate the persistent factors of COVID-19 diagnosis using four scenarios of asymptomatic inclusion. We use Bayesian Logistic Regression to identify the factors of COVID-19 positivity, which can address issues in the traditional approach such as overfitting and uncertainty. This study discovers three main findings: (1) COVID-19 can infect people regardless of age; (2) Among twelve symptoms of coronavirus (COVID-19), five symptoms increase the COVID-19 likelihood, and two symptoms decrease the possibility of COVID-19 infection; and (3) From an epidemiological perspective, the contact history rises the probability of COVID-19, while healthcare workers and people who did travel are less likely to become infected from COVID-19. Therefore given this study, it is essential to be attentive to the people who have the symptoms and contact history. Surprisingly, health care workers and travelers who apply health protocols strictly according to the rules have a low risk of COVID19 infection.