Generating Questions Bank for Adaptive Assessment Using Machine Learning Techniques: Review

Adaptive assessments have redefined educational evaluation by offering personalized and dynamic testing tailored to individual learner abilities. This paper reviews key advancements in adaptive assessment systems, emphasizing the integration of artificial intelligence (AI) and machine learning (ML) to enhance question generation, difficulty prediction, and proficiency estimation. It explores traditional and adaptive frame-works, including Computerized Adaptive Testing (CAT) and psychometric theories like Item Response Theory (IRT). Through a comparative analysis of recent studies, the paper highlights innovative algorithms, challenges in scalability and fairness, and opportunities for future research. The findings underscore the transformative potential of adaptive assessments in creating equitable and efficient educational environments. This review not only synthesizes current advancements but also identifies critical gaps and proposes future research directions for more inclusive and scalable adaptive assessment systems.