Predicting BTC price trends with social media sentiment: leveraging transformers model

• Ano de defesa: 2025
• Autor(a) principal: Hidalgo, Rodrigo Soares de Andrade
• Orientador(a): Colombo, Jefferson Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Aprendizado de máquina
• Palavras-chave em Inglês: Bitcoin; Transformers; Machine learning
• Link de acesso: https://hdl.handle.net/10438/36647

This study explores the prediction of BTC (BTC) price movements using sentiment analysis from social media, combined with advanced machine learning techniques, with a particular focus on the innovative application of Transformers. In a highly volatile financial market sensitive to external influences, understanding the impact of investor sentiment is crucial for predicting price trends. This work employs sentiment data extracted from plat- forms such as Twitter and Reddit, processed through natural language processing (NLP) techniques to categorize sentiment as positive, negative, or neutral. The Transformer model is utilized to develop forecasts on the future direction of BTC prices. Known for its ability to capture complex temporal patterns and dependencies over long sequences, the Transformer enhances prediction accuracy by effectively modeling long-range dependencies in sequential data. The study implements a sensitivity analysis to assess the impact of different sentiment intensities on BTC prices, providing a deeper understanding of market dynamics influenced by sentiment events, such as large ”whale” transactions or significant news. The expected results include improved accuracy in predicting BTC price movements, offering valuable insights for traders and investors. This work contributes to the existing body of research on the use of sentiment analysis in financial markets, particularly in the context of cryptocurrencies, and underscores the potential of Transformers as powerful tools for forecasting market trends based on sentiment data.