PROBABILISTIC FORECASTING MODEL FOR THE COVID-19 PANDEMIC BASED ON THE COMPOSITE MONTE CARLO MODEL INTEGRATED WITH DEEP LEARNING AND FUZZY SYSTEM

摘要

THERE ARE SEVERAL TECHNIQUES TO SUPPORT SIMULATION OF TIME SERIES BEHAVIOR. IN THIS CHAPTER, THE APPROACH WILL BE BASED ON THE COMPOSITE MONTE CARLO (CMC) SIMULATION METHOD. THIS METHOD IS ABLE TO MODEL FUTURE OUTCOMES OF TIME SERIES UNDER ANALYSIS FROM THE AVAILABLE DATA. THE ESTABLISHMENT OF MULTIPLE CORRELATIONS AND CAUSALITY BETWEEN THE DATA ALLOWS MODELING THE VARIABLES AND PROBABILISTIC DISTRIBUTIONS AND SUBSEQUENTLY OBTAINING ALSO PROBABILISTIC RESULTS FOR TIME SERIES FORECASTING. TO IMPROVE THE PREDICTOR EFFICIENCY, COMPUTATIONAL INTELLIGENCE TECHNIQUES ARE PROPOSED, INCLUDING A FUZZY INFERENCE SYSTEM AND AN ARTIFICIAL NEURAL NETWORK ARCHITECTURE. THIS TYPE OF MODEL IS SUITABLE TO BE CONSIDERED NOT ONLY FOR THE DISEASE MONITORING AND COMPARTMENTAL CLASSES, BUT ALSO FOR MANAGERIAL DATA SUCH AS CLINICAL RESOURCES, MEDICAL AND HEALTH TEAM ALLOCATION, AND BED MANAGEMENT, WHICH ARE DATA RELATED TO COMPLEX DECISION-MAKING CHALLENGES.