With the rapid advancement of artificial intelligence and smart education, academic performance prediction has emerged as a critical research direction within educational data mining. This paper provides a systematic review of recent progress in this field, focusing on technological evolution, application scenarios, model adaptation, emerging trends, and persistent challenges. The development of prediction technologies has progressed through three stages: statistical methods, machine learning, and deep learning. These approaches have been widely applied to grade prediction, academic risk and dropout prediction, knowledge tracing, and personalized learning recommendation. Different models var... More >