AI-Powered Sentiment Analysis for Financial Forecasting through Tweets

This Capstone Project presents an approach to financial forecasting by leveraging sentiment analysis on social media content. Utilizing AI models from Hugging Face known as DistilRoBERT a, more specifically DistilRoBERT a-financial-sentiment, an AI community building the future, I examined and analyzed the sentiments of over 10,000 statements extracted from X (formerly Twitter), uncovering patterns that could detect trends of the market.

The chosen model was selected for its fine tuned performance in natural language processing and effectiveness for sentiment classification. After collecting a substantial number of statements from X (formerly Twitter) and analyzing them, it resulted in correlations of market trends and collective sentiment. These findings suggest that sentiment analysis, powered by AI, can provide very valuable insights into public sentiment and the fluctuations of the financial market.

The project not only proves useful to this use case, but also demonstrates the practical application of AI in extracting meaningful patterns from vast, unstructured datasets.