A Survey on Ensemble Learning Approaches for Classification of Real Time COVID-19 Data and Forecasting Severity of the Epidemic

EasyChair Preprint 7439

Abstract

Healthcare facility at the side of support of recent technologies like analysis using Artificial intelligence and Machine learning algorithms play a very important role to defend against in prediction, diagnosis and treatment of novel corona virus disease which emerged in Wuhan City and speedily spreading throughout the globe. In this paper we review the Ensemble learning approaches for classifying and forecasting COVID-19 epidemic which will further help to take preventive steps and necessary actions. For the welfare of human race, in this paper we propose to utilize the benefits of ensemble approach to understand the behavior and pattern of corona virus in order to accurately predict the expected new cases of COVID 19 by using the current and updated streams of information. Review of latest related literature including techniques like AI, Machine learning, COVID -19, ensemble approaches etc. The latest information regarding different prediction techniques and analysis methods of COVID 19 data using machine learning, AI and ensemble techniques by different authors were collected and analyzed to identify the role of machine learning and AI for prediction and diagnosis of COVID 19 disease. In this paper, we have used the Ensemble approaches to predict the total active cases, recovered cases, and death cases all over the world. It is implemented using the python library “sklearn”. The prediction can further assist to take the necessary decisions related to lockdown period, medical facilities, government policies and rules.

Keyphrases: Artificial Intelligence, COVID-19, Corona Virus, Predictive Analysis, ensemble learning, machine learning

@booklet{EasyChair:7439,
  author    = {Monika Arya and Chaitali Choudhary},
  title     = {A Survey on Ensemble Learning Approaches for Classification of Real Time COVID-19 Data and Forecasting Severity of the Epidemic},
  howpublished = {EasyChair Preprint 7439},
  year      = {EasyChair, 2022}}