Composite Monte Carlo decision making under high uncertainty of novel coronavirus epidemic using hybridized deep learning and fuzzy rule induction

Area of Science:

• Epidemiology
• Computer Science
• Data Science

Background:

• The novel coronavirus epidemic presents significant decision-making challenges due to high uncertainty and limited early data.
• Traditional forecasting methods like Composite Monte-Carlo (CMC) simulations rely on data quality and statistical distribution assumptions, which can be limiting in dynamic epidemic scenarios.

Purpose of the Study:

• To enhance Composite Monte-Carlo (CMC) simulations for improved epidemic forecasting.
• To provide decision-makers with more reliable stochastic insights into epidemic development.
• To explore the integration of deep learning and fuzzy rule induction for more accurate epidemic modeling.

Main Methods:

• A case study employing a deep learning-enhanced CMC simulation.
• Integration of fuzzy rule induction techniques with CMC.
• Utilizing temporal-spatial data, population density, travel mobility, medical resources, and quarantine timeliness as input factors.

Main Results:

• The deep learning-based CMC, combined with fuzzy rule induction, offers improved fitted outputs compared to simplistic Monte Carlo (MC) assumptions.
• Min-max rules derived from fuzzy logic provide foresight into the extreme ranges of future epidemic possibilities.
• Enhanced stochastic insights aid decision-makers in navigating epidemic uncertainty.

Conclusions:

• The integration of deep learning and fuzzy rule induction significantly improves the reliability and accuracy of CMC simulations for epidemic forecasting.
• This advanced approach offers valuable tools for public health decision-making during epidemics.
• The methodology provides a more robust framework for understanding and predicting epidemic trajectories under uncertainty.