COVID-19 Prediction and Detection using Machine Learning and Artificial Intelligence

Abstract

Abstract: This study focuses on the use of Artificial Intelligence and Machine Learning models to predict the possible corona positive cases and detect the spread of corona among individuals by scanning their X-Rays. Linear Regression Model is used to achieve the desired Prediction output. Along with Linear Regression we have used SVR model to attain comparative analysis among different model prediction techniques. Finally a forecasting model Holt’s model was developed which seemed to be better than the above two models. Classical ML related Python libraries are extensively used in this project to train and preprocess large amounts of data. In the Detection phase, the Convolutional Neural Network model is used to detect COVID-19 infection among individuals. In the detection phase we have added multiple filtering rounds to achieve the best accuracy. The model is trained upto certain epochs to improve the accuracy and reduce the overall loss. The datasets used for all the above-mentioned models are extracted from github repositories associated with some data collection organisations. This study also includes the global analysis of COVID-19 and displays the trend shown in a graphical representation. Keywords: X-Rays, Linear Regression, SVR, Holt’s Model, Convolutional Neural Network, epochs