Browsing by Author "Maria Farooq"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    COVID Misinformation Detection on Social Media Using Deep Learning Technique
    (UMT, Lahore, 2023) Maria Farooq
    The COVID-19 epidemic has prompted a surge in the increase of misinformation on social media platforms, which can lead to dangerous consequences. Unfortunately, not all of this information is accurate or trustworthy, and misinformation about the virus and its impact has become widespread. This rapid increase in misinformation harms people living in society. Misinformation is particularly predominant in the progressing COVID pandemic, prompting people to tolerate bogus and possibly injurious cases and articles. Therefore, detecting and identifying such misinformation on social media is critical. In recent years, there have been significant efforts to develop automated tools for detecting misinformation on social media. These tools use various techniques, such as machine learning, natural language processing, and network analysis, to identify patterns and characteristics of misinformation. Our research explored using deep learning models, specifically BERT and Distilbert, for COVID-19 misinformation recognition on social media. We utilized a large COVID misinformation English tweet dataset from Kaggle. We then trained and evaluated the models on this dataset, achieving high accuracy and F1-score. Bert achieved 86% accuracy, and Distilbert achieved 98% accuracy. The ascertainment of the proposed framework represents better achievement than the other model's accuracy, recall, precision, and F-score. Our results suggest that BERT and Distilbert effectively identify COVID-19 misinformation on social media and can be utilized to support public health initiatives in combating the spread of harmful misinformation.