LSTM-RNN Based Predictive Stock Analysis with Real-Time Sentiment Interpretation

This paper presents a real-time AI-powered dashboard for stock market analysis that integrates advanced deep learning models and sentiment analysis to assist investors with data-driven insights. The system leverages APIs such as Yahoo Finance and Finnhub to fetch live and historical stock data, and NewsAPI to gather current financial news. Sentiment scores are computed using the VADER (Valence Aware Dictionary and Sentiment Reasoner) tool, while Google’s Gemini AI enhances interpretation by identifying nuanced market signals. Technical indicators like RSI, MACD, ADX, and Bollinger Bands are used for trend analysis. Machine learning models including Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), and Gated Recurrent Unit (GRU) are employed to predict future price movements by learning from historical data patterns. Built with the Streamlit framework, the dashboard offers an interactive, secure, and user-friendly interface. This multi-modal approach—combining real-time data, NLP-driven sentiment analysis, and AI-based forecasting—aims to reduce investor uncertainty and improve decision-making in volatile market conditions.