Sentiment analysis & machine learning model to predict stock movement. Experiment with various technique to generate trading signal such as hybrid Simple Moving Average combine with sentiment analysis and 5 machine learning technique (Logistic Regression, Naive Bayes, SVM, Random Forest & MLP). Dataset was mining using web scraping technique & API.
SVM | Random Forest | MLP
0.80 Sharpe Ratio on SMA & Sentiment Analysis | 26.2% Annual Return
0.79 Sharpe Ratio on MLP Model | 42.6% Annual Return