Article’s

Stock market prediction remains a challenging task due to the highly volatile, non-linear, and dynamic behavior of financial markets, where price movements are influenced by both historical trends and external factors such as news and investor sentiment. This paper presents a hybrid data-driven framework that combines Bidirectional Gated Recurrent Units (BiGRU) and Bidirectional Long Short-Term Memory (BiLSTM) networks for accurate stock price prediction, along with Natural Language Processing (NLP)-based sentiment analysis of financial news. The BiLSTM component is designed to capture long-term dependencies in time-series data, while the BiGRU enhances computational efficiency and learns short-term patterns effectively. The bidirectional structure enables the model to leverage contextual information from both past and future sequences, improving predictive performance. Simultaneously, financial news data is collected and processed using NLP techniques, including text preprocessing and sentiment extraction, to quantify market sentiment. These sentiment scores are integrated with historical stock prices and technical indicators to provide enriched input to the hybrid model. Experimental results indicate that the proposed approach significantly improves prediction accuracy and reduces forecasting errors compared to traditional models. The integration of sentiment-aware features enhances the model’s robustness, making it suitable for real-time decision support in dynamic financial environments.