Predicting COVID-19 Outbreaks: Leveraging Machine Learning and Deep Learning Models for Trend Analysis

Machine learning plays a significant role in the healthcare domain. It is widely categorized as supervised, unsupervised, and reinforcement learning. This paper aims to give a detailed description of our experimental research work, where we used various supervised machine learning algorithms to develop a prediction model for outbreaks of the novel Coronavirus that has spread worldwide and caused many deaths. In the history of the world, it is one of the most disastrous Pandemics. This work aims to understand how the machine learning, an ensemble, and the deep learning models work in the real-time dataset. Further, we analyzed the current trend and pattern of the COVID-19 outbreaks. Then, we predicted future confirmed or new cases by training the past COVID-19 dataset using machine learning algorithms such as Linear Regression, Polynomial Regression, K-Nearest Neighbor, Decision Tree, Support Vector Machine, and Random Forest. The Decision Tree and the Random Forest algorithms perform better than SVR for our experimental dataset. Generally, the SVR performs poorly in separating the data using the hyperplane. It primarily gives a lower performance in this problem. Ensemble (Voting, Bagging, and Stacking) and deep learning model (ANN) also predict well. After the prediction, we evaluated the model using MAE, MSE, RMSE, and MAPE.