Assessing the impact of distance education contents on learning success using machine learning

This study examines the impact of distance education course contents (e.g., videos, forums, documents) on student success using machine learning. The dataset, collected from Akdeniz University during the 2019–2020 Spring semester, includes content usage metrics and success grades. Six machine learning models—Decision Tree, Neural Network, Naive Bayes, Logistic Regression, Random Forest, and XGBoost—were applied to analyze the relationship between engagement and academic outcomes. Success scores were categorized into three datasets for classification. The results showed that XGBoost achieved the highest testing accuracy (97%), followed by Random Forest (94.43%). The findings highlight that active engagement with specific content types, particularly videos and forums, significantly predicts learning outcomes. The study provides empirical evidence for instructional designers to optimize distance education materials for better student success.