Water quality research plays a pivotal role in addressing and mitigating water pollution issues, while accurate water quality forecasting is vital for safeguarding the ecological integrity of watersheds. In recent years, the integration of machine learning models into water quality prediction has garnered significant attention from scholars due to their inherent advantages, including adaptability, self-learning capabilities, high efficiency, and fault tolerance. This paper aims to provide a comprehensive overview of the application of machine learning techniques in water quality prediction models. Additionally, it critically examines the existing challenges and limitations associated with these prediction models. Furthermore, this study presents a forward-looking perspective on the potential utilization of machine learning approaches in forecasting water quality models specific to the Erhai Lake region. By consolidating and analyzing the available knowledge, this research endeavors to contribute to the advancement of machine learning-based water quality prediction methodologies in order to enhance the effectiveness and accuracy of future predictions in the Erhai Lake area.