Urban growth prediction is crucial in contemporary urban planning and sustainable development, driven largely by the increasing availability of data and advancements in Machine Learning (ML). This literature review examines current research trends in ML-based urban growth prediction, categorizing methodologies, data sources, accuracy assessments, and limitations. Traditional approaches like linear regression are contrasted with hybrid models such as Artificial Neural Networks (ANN) and ensemble methods, highlighting their superior predictive capabilities when integrating remote sensing data, Geographic Information Systems (GIS), and socio-economic variables. While traditional methods show limited application, hybrid models demonstrate enhanced robustness and accuracy by combining multiple ML techniques. Challenges identified include data quality, consistency, model interpretability, and the urgency for real-time prediction capabilities. The review underscores the gaps in current research, including the need for comprehensive comparative studies and the application of deep learning architectures like convolutional neural networks. By leveraging interdisciplinary approaches, urban growth prediction can advance towards more effective and sustainable urban development strategies. Continued advancements in ML and data integration are crucial for addressing the complexities of urban growth dynamics and enhancing the resilience of cities in an era of rapid urbanization.