Bayesian inference and machine learning applications in health care problem
Robust and accurate inference and decision making are challenging due to uncertainty, preference, and high dimensionality. The problem becomes more challenging with limited information. The aim of this project is to make a scalable and accurate inference to make better decisions and applying the concept in the real-world environment.
Human cultural dimensions and behavior during COVID-19: a Bayesian approach
- COVID-19 pandemic is the enormous crisis the world faces, costing 4.5 million lives in 2 years. Almost every country implemented both pharmaceutical and Non-Pharmaceutical Interventions (NPI) (restriction) to control the pandemic. However, the restriction helps to alleviate suffering and death from pandemics but raises socio-economic hardship. Economic disruptions begin as most countries declare social distancing, lockdown, travel restriction, non-essential business closure. Social trouble arises as the government impels people in the house, children cannot go to the school and recreation center, and people from all sectors are confined. Human behavior decision-making plays a crucial role, especially NPI, in delimiting the pandemic; it varies from culture to culture. The research aims to show the importance of human behavior for practices social distancing to control the pandemic. Integrating evolutionary game theory with our disease model, we develop a conceptual framework and deterministic model. The Bayesian melding approach is adopted to answer the research question. We specify the basic reproduction number using the Bayesian inference technique to estimate the disease model’s parameter. This research will help to highlight the importance of human choice as an influence in disease modeling and pandemic control to defend current and future uncertain diseases and pandemics.