Stock market prediction using News API and deep learning algorithms

FinWizard is a sophisticated stock prediction system based on deep learning which incorporates real-time data from Yahoo Finance alongside sentiment analysis from financial headlines utilizing the valence aware dictionary and sentiment reasoner (VADER) tool. Unlike conventional models which use historical prices as the only input, FinWizard attempts to improve accuracy in key sectors like banking, FMCG, Pharma, IT, and automotive by blending quantitative and qualitative data. Employing long short-term memory networks, the model issues buy, sell, or hold signals and outperform other architectures like convolutional neural networks , recurrent neural networks, and hybrid ensembles. Evaluated on mean absolute percentage error (MAPE), Mean absolute error (MAE) and root mean squared error (RMSE)„ the LSTM model has consistently outperformed benchmarks achieving MAPE scores below 2%, with the banking and healthcare sectors yielding the highest accuracy. The system responds to changes in market volatility by combining moving averages and other technical indicators with sentiment-driven signals to provide situational context-relevant trading recommendations. It is also able to provide classification based on sectors as well as market capitalization, broadening its applicability. Fin-Wizard has proven its reliability and flexibility across different financial conditions, thus, is suitable for short-term retail trading and long-term institutional strategies. By combining market and sentiment analysis, FinWizard offers a real-time and comprehensive decision support system.