A Novel Ensemble-Based Sentiment Classifier for Deep Fake Tweets Using Majority Voting With BERT

This study affords a novel approach to sentiment analysis of deepfake tweets by way of growing a sentiment majority voting classifier combined with transfer mastering-based function engineering. The goal is to appropriately distinguish between human-generated and robotic-generated tweets, leveraging the electricity of gadget getting to know and natural language processing. A two-column dataset containing the tweet text and corresponding labels (human or robotic) serves because the input. We appoint Bidirectional Encoder Representations from Transformers for function extraction, taking gain of its superior contextual understanding of text. The proposed version integrates the strengths of each BERT providing a green and scalable answer for deepfake tweet detection. This approach contributes to more desirable detection accuracy, presenting a practical framework for figuring out automated content in social media environments.