A Hybrid ML Algorithm Trading Model combining Sentiment, Technical Indicator and Risk Management

Large language models (LLMs) and sentiment-aware machine learning are examples of technological developments that have influenced algorithmic trading using non-traditional data sources such as social media. The unpredictable, non-stationary nature of financial data makes stock market prediction difficult. Machine learning has emerged as a powerful paradigm for dynamic market adaptation. The machine learning model’s ability to reduce trading risk and maximise profits has been limited by the current methodologies’ frequent lack of regard for structured risk management. In this paper, we propose a hybrid model combining sentiment analysis and technical indicators along with market data to predict the trend in the market. We also explore the returns obtained from the model and compare them with our risk strategy and traditional buy-and-hold (B&H) strategy. The results show that machine learning can decisively outperform traditional methods if paired with efficient risk management, which contributes to our understanding of machine learning-driven trading.