Advancing Personalized and Adaptive Learning Experience in Education with Artificial Intelligence

The challenge for today's learning systems is to provide effective access to knowledge and contents that are well-relevant to learners' background and interest levels. Majority of personalized educational platforms lack methods to effectively support the needs of learners who are generally heterogeneous in terms of intellectual abilities, learning pace, preferences, academic background, etc. Hence, there is a need to provide powerful mechanisms to organize such learning and educational activities and to adapt best pedagogical decisions to the needs of each learner. In this work, we addressed major challenges of adaptive and personalized learning that have been neglected in prior studies. To this aim, we leverage effective Supervised Machine Learning (ML) techniques to adaptively schedule assignments and educational activities based on the students' needs, preferences, and background. The proposed intelligent system is trained based upon different academic factors from student learners' characteristics such as proficiency l evel, i nterest level, remote/in-person preference, and assignment type preference and prescribes a proper learning plan to maximize the students' overall grade and satisfaction rate at the end of course. In addition, we conduct analysis of demographic parameters such as gender and race and their effects on students' success and academic satisfaction. For a comprehensive analysis, five different ML models including Logistic Regression (LR), K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) are examined. The experimental results demonstrate the superior effectiveness of the Random Forest classifier in comparison to other ML algorithms. The proposed intelligent system based on RF model achieves a 94% F1-score and accuracy rates, enabling accurate assignment of the most effective learning mode out of the four available options for learners.