A A Dynamic Mining of Interesting Web Usage Patterns for Personalized E-Learning: A Systematic Review
Personalized e-learning adapts educational materials and learning paths to individual learner characteristics. This adaptation is often supported by analyzing learner interactions through Web Usage Mining. The present survey reviews research from 2010 to 2025 that addresses dynamic web usage pattern mining for personalized e-learning environments. Four main contributions stand out. First, a hierarchical classification organizes approaches across five interconnected levels: data sources, preprocessing, pattern discovery, dynamic mining, and recommendation systems. A comparison then follows between static and dynamic mining approaches—tracing temporal and incremental methodologies, something earlier surveys often overlooked. Third, the analysis turns to recommendation mechanisms: collaborative filtering and content-based methods. Finally, ongoing challenges are identified and future directions suggested. Static algorithms such as Apriori and FP-Growth, according to the review’s conclusion, are increasingly giving way to dynamic techniques—incremental mining and sliding window analysis, for instance. Dynamic methods capture the evolution of learner behavior more effectively. Several research gaps remain. Despite their proven effectiveness, deploying these dynamic techniques in real-world educational settings continues to face obstacles, with scalability and latency being the most prominent. More critically, metrics for evaluating pattern interestingness based on educational impact remain underdeveloped.
Metrics
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
