Date of Publication : March 2026

Abstract: The stock market is inherently volatile and influenced by numerous dynamic factors, including public sentiment. Traditional forecasting models rely solely on historical price data, often overlooking the qualitative impact of market sentiment. This project proposes a hybrid predictive system that integrates Long Short-Term Memory (LSTM) neural networks with sentiment analysis to enhance short-term stock price forecasting. The model analyzes historical stock data along with sentiment extracted from news articles and social media using natural language processing (NLP) techniques. A web-based interface, built with React.js, enables users to interact with the system, while Firebase Firestore serves as the cloud-based backend for real-time data storage and access. The platform is designed with a focus on usability for beginner investors, aiming to bridge the gap between complex market analytics and intuitive decision-making tools.

Reference :