Study on the influencing factors of online learning effect based on decision tree and recursive feature elimination

摘要: With the popularity of online learning, more and researchers have attached great importance to relationship between learning effect influence factors in courses. In literature works, Logistic Stepwise Regression algorithm is most used method. But this method has limitation run time especially when dimension data large. Besides that, it can't rank factors. Aiming at above shortcomings, paper proposes a novel approach analyze influencing which based on combination decision tree recursive feature elimination. Firstly, sorting conduct preliminary screening, form candidate set. Then, elimination features by their importance. At stage, (LR), Support Vector Machine (SVM) Decision Tree (DT) models are separately obtain each collating sequence By averaging these sequences, final ranking achieved. Finally, an experiment carried out Open University Learning Analytics dataset, results show that behavior important impact effect. Positive behaviors can lead better