Hyper Capsule LSTM-Gated GAN with Bayesian Optimized SVM for Cloud-based Stock Market Price Prediction in Big Data Environments

In the modern era, big data is a brand-new and developing buzzword. With a significant expansion of finance and business growth and forecast, the stock market is a dynamic, ever-evolving, unpredictable, and fascinatingly promising specialty. This study presents a novel approach for enhancing forecast accuracy through optimal feature selection combined with deep learning techniques. By employing an Artificial intelligence method to identify and select the most significant features influencing stock prices, we mitigate the risks of overfitting and improve model interpretability. To propose an advanced methodology called Hyper Capsule LSTM Gated Generative Adverbial Network (HCG-GAN) with Bayesian Optimized Support Vector Machine (BOSVM) for stock market price prediction, which is well-suited for time-series data. A comparative analysis is conducted to evaluate the performance of our model against traditional prediction methods. The preliminary process takes place in stock market pricing data log normalization using a Min-max z-score normalizer. Then Active stock distinction impact rate (ASDIR) is estimated to find the scaling factor of stock market mean changes. The prediction performance of the proposed model is compared with that of the benchmark models CNN-LSTM, DLSTMNN, and ANN-RF using evaluation metrics of accuracy, precision, recall, F1-score, AUC-ROC, PR-AUC, and MCC. Results indicate that the integration of optimal feature selection not only boosts prediction accuracy but also ensures robustness against market volatility. This work contributes to the growing body of literature on artificial intelligence applications in finance, offering insights that can significantly enhance trading strategies and investment decisions.