Construction of an Online Learning Resource Recommendation Model Based on Artificial Raindrop Algorithm in The Context of Smart Education

Abstract

The explosive growth of online education has transformed traditional teaching, resulting in difficulties such as course selection, complex learning paths, and information overload. This study proposes an online resource recommendation (RR) model based on the artificial raindrop algorithm (ARA) to address these issues. In smart education (SE), the study first establishes a learner model (LM) to capture learners’ personalized needs and characteristics. The ARA is then used to construct an online learning RR model, simulating raindrops’ search to find the optimal learning resources. A perturbation mechanism is introduced to improve the algorithm’s diversity and search ability, enhancing the quality of recommendations. Experiments showed the proposed algorithm’s training time of 3.605 seconds, shorter than comparison algorithms, and better accuracy, recall, and F1 scores of 0.9531, 0.07639, and 0.1272, respectively. This study offers new methods and ideas for improving course selection and learning paths in online education platforms.