Comparative Analysis of Artificial Intelligence Models in Stock Security Index Forecasting with Sentiment Analysis of Economic News

AI-based models are valuable financial tools because they generate precise predictions and mitigate investment risks. This aids investors in making judgements based on factual infor-mation. It is essential to monitor and analyse various forecasting methods in the financial market, both present and future, with profitability as a critical factor. The studies incorporated in this analysis were sourced from three prominent academic databases and were published from 2019 to 2024. The selected research employed a three-stage approach of preparation, execution, and analysis. The study investigates technical evaluation, profitability analysis, hybrid modelling, and model output. To assess the study's methodological strength, 10 Likert-type quality criteria questions were used to evaluate the articles. Establish inclusion and exclusion criteria to initiate the examination. An increasing number of organisations are utilising ensemble and hybrid models that integrate LSTM and SVM to forecast market movements and prices. Hybrid models that employ AI algorithms for feature engineering exhibit considerable potential, similar to ensemble methods. Performance-based research excludes profitability measures, investment strategies, and real-time or simulated trading. The financial forecasting literature needs to be revised, and future studies could concentrate on multi-class or output-related aspects.