Big Data Analysis and prediction of COVID-2019 Epidemic Using Machine Learning Models in Healthcare Sector

Abstract

A global pandemic of COVID-19 is underway. The world now thinks coronaviruses cause all illnesses, including this pandemic. The world is also seeing fast viral spread. COVID-19 spreads swiftly by respiratory droplets and contact, according to the WHO. Big data analytics technologies are essential for developing the information needed to make choices and take preventative action. The large volume of COVID-19 data available by multiple sources necessitates, however, a review of big data analysis's roles in containing the virus, highlighting the primary obstacles and future directions in COVID-19 data analysis, and offering a framework for relevant current applications and studies to aid in COVID-19 analysis research in the future. The aim of ML, DL methods in COVID-19 research and applications is the focus of this study. This endeavor use machine learning to evaluate and predict COVID-19 pandemic behavior to help control a pandemic. A study uses the Israeli Department of Health dataset to predict COVID-19 results using ML and big data analytics. The test results were analyzed using CNN-LSTM and Decision Tree classification algorithms. Model efficacy was measured by F1-score, recall, accuracy, and precision. CNN-LSTM was the most accurate, with 96.34% accuracy and strong predictive ability. Also, for the model explainability used LIME model. The findings may be used by various government agencies to implement remedial actions. It may be simpler to fight COVID-19 if methods for infectious illness forecasting were readily available.