ACCURATE STOCK MARKET HIERARCHICAL TREND PREDICTION FORECASTING WITH ARTIFICIAL OPTIMIZATION ALGORITHM
Abstract
The landscape of stock market prediction is undergoing a profound transformation driven by technological advancements and data-driven methodologies. Within this shifting paradigm, artificial intelligence (AI), particularly deep learning (DL) is emerging as a transformative tool to enhance predictive accuracy stock market trend, a textual embedding construction method is proposed to encode multiple textual features, In recent years machine learning and deep learning have become popular methods for financial data analysis, including financial textual data, numerical data, and graphical data. An optimized deep LSTM network with the ARO model is created to predict stock prices. DJIA index stocks are used as the dataset. LSTM-ARO is compared with one artificial neural network (ANN) model, three different LSTM models, and LSTM optimized by Genetic Algorithm (GA) model. New hybrid model, termed Hierarchical Decomposition-based Forecasting Model (HDFM), to decompose and forecast stock prices in a hierarchical fashion. The first combined sub-series is subjected to a second decomposition with variational mode decomposition (VMD). To enhance sentiment analysis accuracy, this study proposes text classification using Bidirectional Encoder Representations from Transformers (BERT) and its variants for natural language processing. Experimental results demonstrate the effectiveness of combining BERT with Convolutional Neural Networks (CNN). The purposedesign philosophy ensemble deep learning technologies future stock price trends more effectively and can better assist investors in making the right investment decision than other traditional methods.