Artificial intelligence in adaptive education: a systematic review of techniques for personalized learning

Artificial Intelligence (AI) has significantly transformed digital education by enabling personalized and data-driven learning experiences. By incorporating machine learning (ML), deep learning (DL), and multimodal analytics, AI systems adapt instructional content to match individual learner profiles in real time. This systematic review examines AI-powered adaptive learning technologies with an emphasis on supervised and unsupervised learning, reinforcement learning (RL), and multimodal data integration. It assesses how these technologies enhance personalization, boost learner engagement, and promote educational equity. Adhering to PRISMA guidelines, this review analyzes 142 peer-reviewed empirical studies published between 2015 and 2025, sourced from databases including Scopus, Web of Science, and IEEE Xplore. Studies were selected based on their application of AI techniques in personalized digital learning environmen. Supervised learning models such as support vector machines (SVMs) and decision trees support learner classification and performance prediction. Unsupervised clustering techniques uncover behavioral patterns, while reinforcement learning enables dynamic content sequencing to foster learner autonomy. Multimodal approaches that integrate affective and biometric data further refine personalization. Despite these advances, critical challenges persist in model interpretability, data privacy, and scalability, especially in low-resource settings. This review highlights the expanding influence of AI in personalized education and outlines future research directions, including the use of explainable AI (XAI), modular system designs, and culturally responsive frameworks. These advancements are vital for developing ethical, inclusive, and scalable learning systems.