Infodemic Vs Pandemic: Covid-19 Crisis in Pakistan Due to Rumors on Twitter

The global spread of the Coronavirus disease 2019 (COVID-19) led to a pandemic that threatened millions of lives. Infodemics have emerged as a significant area of concern. Today, social media platforms play a crucial role in reflecting public sentiments and opinions on current events. Twitter, in particular, has gained considerable attention from researchers for analyzing public sentiment. Predicting pandemics and infodemics based on public sentiment expressed on Twitter has become a fascinating research area. This study focuses on users in Pakistan with the highest number of followers who have posted the most tweets related to COVID-19. The manually annotated dataset includes 2,000 tweets from 1,000 users for training and 380 tweets for testing, collected between June and July 2020. Data processing involved manual labeling and feature extraction with the help of SentiWordNet. The proposed model, KORONV, gathers tweets containing COVID-19-related hashtags. Various machine learning algorithms were applied, with Long Short-Term Memory (LSTM) achieving the highest accuracy of 98%. These methods will be used to detect patterns based on existing algorithms and datasets, ultimately providing solutions for identifying and classifying tweets as positive, negative, or neutral.