Volume 2 - Issue 3, May - June 2026

The stock market is a highly volatile environment in which accurate prediction remains a formidable challenge due to the multitude of simultaneous influencing factors. This paper presents a hybrid system that combines Twitter sentiment analysis with Long Short-Term Memory (LSTM) networks to predict the next-day closing values of publicly traded stocks. The proposed approach exploits the temporal correlation between public opinion and market movement by applying Part-of-Speech (POS) tagging for sentiment polarity classification and Random Forest for text-based news classification, achieving accuracy exceeding 80%. A web-based prediction interface allows users to submit news text and receive instant positive or negative sentiment labels. Experimental results demonstrate that the system reliably captures sentiment-driven market signals from social media and financial news, offering a practical and scalable tool for retail investors.

G Kasi Reddy, Pallapu Monika, P Mithra Reddy, Jangam Ashwik,"Stock Market Sentiment Analysis and Prediction Using Hybrid LSTM and NLP Approach" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(3): Page(609-613) May-June 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.