Description

COVID-19 is a life threatening disease which has an enormous global impact. As the cause of the disease is a novel coronavirus whose gene information is unknown, drugs and vaccines are yet to be found. The used dataset is extracted from twitter provided by the UCI Repository. After gathering the tweets, the dataset is cleaned using the preprocessing techniques, they are labelled and a model is developed which is effective for detecting the actual sentiment behind a tweets related to COVID-19. The contribution of this works is the performance evaluation of various machine learning classifiers using proposed feature set. Tweets are classified as extremely positive, neutral, extremely negative, positive or negative using machine learning algorithm (Logistic Regression) and the deep learning algorithm (CNN). Performance of classifiers is evaluated on the accuracy, precision, recall and F1-score.