This repository contains the complete code for my dissertation project titled "Hybrid Machine Learning Model for Algorithmic Trading Using Technical, Fundamental, and Sentiment Data." The project develops a sophisticated hybrid machine learning model that integrates three key types of stock market data: technical indicators, fundamental analysis, and sentiment analysis from social media and news sources.
The core objective of this project is to enhance the accuracy and timeliness of stock market predictions. It employs advanced machine learning techniques, including Random Forest, XGBoost, and LSTM networks, to create predictive models that significantly outperform traditional single-dataset models. This repository includes all datasets used, preprocessing scripts, model training files, and the code for an automated trading bot that utilizes the model's predictions to execute trades.
This work demonstrates the potential of combining diverse data sources in predictive modeling for financial markets, providing a comprehensive framework that can be expanded in future research.