Forecasting Bitcoin Price Direction from Twitter Sentiment and Market Data

This research explores how social media sentiment affects cryptocurrency price movements using the 17.7 million Bitcoin Tweets dataset from Kaggle. By applying VADER sentiment analysis and training Random Forest and XGBoost models, the study achieved 74.15% accuracy, highlighting sentiment and market features as key predictors.

Forecasting Bitcoin Price Direction from Twitter Sentiment and Market Data, explores how social media emotions influence cryptocurrency prices. Using the 17.7 million Bitcoin Tweets dataset from Kaggle (2017–2019), I applied VADER sentiment analysis to link tweet tone with hourly market data. After cleaning, visualizing, and preprocessing the data, I trained two machine learning models—Random Forest and XGBoost—to predict Bitcoin’s price direction. XGBoost performed best, reaching 74.15% accuracy and showing that both market and sentiment features (like Compound Score and tweet volume) were key predictors. Future plans include feature engineering, hyperparameter tuning, and combining models for improved accuracy.

Explore More!

Research Paper