This study conducts a systematic literature review to examine the evolution and application of machine learning (ML), deep learning (DL), and hybrid models in stock market forecasting in emerging markets. A total of 100 peer-reviewed journal articles published between 2022 and 2025 were retrieved from the Scopus and Web of Science databases and analyzed. This review explores publication trends, methodological developments, regional research focuses, and the interplay between forecasting models and country-specific contexts. The results indicate a sharp increase in publications over the past two years, with Random Forest and LSTM emerging as the most frequently used techniques. Hybrid models that integrate multiple algorithms have also gained substantial traction, reflecting the need for enhanced predictive performance in volatile markets. However, the analysis revealed a strong geographical skew, with India and China dominating the research landscape and African and Latin American countries being significantly underrepresented. The findings highlight both the growing methodological sophistication of financial forecasting and the critical need for broader geographic inclusivity. This review provides a comprehensive foundation for future research by identifying key gaps, recommending the adoption of interpretable AI, and advocating for standardized performance metrics across forecasting studies in emerging market contexts.