Data mining techniques in blended learning: Prediction of students learning outcomes from online behavior data.

• In Event: Exploring student learning outcomes

Short Description

This study will employ an educational data mining (EDM) method of using online behavior data from the Learning Management System, which is based on the students' interaction activity in the Learning Management System (LMS). Two EDM methods, namely clustering and classifying method will be used to describe and predict students’ level of engagement in the course and its correlation to their learning outcomes. The K-means clustering algorithm will be used to categorize the online behaviors into 3 distinct activity levels; low, moderate, and active. Based on the activity levels, classification algorithms will be used to train the data and later predict students learning outcomes from their online behavior.

Author

• Presenter: Mohamad Ali Baba, Southern Illinois University Carbondale