This systematic review critically examines post-pandemic advancements in machine learning (ML) and deep learning (DL) techniques for stock market analysis. The study, structured around three key themes—Predictive Modelling and Forecasting Techniques, Sentiment Analysis and Natural Language Processing (NLP), and Deep Learning and Advanced Computing in Finance—analyzes recent scholarly contributions between 2020 and 2025. Drawing from 90 open-access peer-reviewed articles identified through a rigorous PRISMA framework, the review highlights how cutting-edge models such as LSTM, CNN, GRU, and reinforcement learning have significantly enhanced stock price forecasting, portfolio optimization, risk assessment, and algorithmic trading. The integration of NLP and sentiment analysis tools has further refined predictive accuracy by extracting market-relevant signals from textual data sources including financial reports, news, and social media. The paper also emphasizes emerging innovations in hybrid and attention-based architectures and domain-specific applications, offering deeper insights into investor behavior and market dynamics. Despite these advancements, challenges such as data quality, model interpretability, and computational demands persist. The study concludes by outlining future research directions in explainable AI, multimodal data integration, and addressing ethical concerns like algorithmic bias and data privacy, ensuring responsible AI adoption in financial markets. This paper contributes to SDG 9 i.e. Industry, innovation and Infrastructure.