Description
Predicting student outcomes early in a learning series can allow for changes in the learning activities to adapt to learner needs and improve outcomes. However, in instructor lead activities, instructors are often faced with a large number of learners and little readily available data on student progress. Particularly, in the case of MOOC's, student data can overwhelm manual human interpretation. Further, in computer driven tutorials, systems have little ability to adapt to students behaviors. This paper reports on an exploratory study of a machine learning system that predicts student grades based on the combination of behavioral and traditional data.
Recommended Citation
Walsh, Kenneth R. and Mahesh, Sathiadev, "Exploratory Study Using Machine Learning to make Early Predictions of Student Outcomes" (2017). AMCIS 2017 Proceedings. 23.
https://aisel.aisnet.org/amcis2017/DataScience/Presentations/23
Exploratory Study Using Machine Learning to make Early Predictions of Student Outcomes
Predicting student outcomes early in a learning series can allow for changes in the learning activities to adapt to learner needs and improve outcomes. However, in instructor lead activities, instructors are often faced with a large number of learners and little readily available data on student progress. Particularly, in the case of MOOC's, student data can overwhelm manual human interpretation. Further, in computer driven tutorials, systems have little ability to adapt to students behaviors. This paper reports on an exploratory study of a machine learning system that predicts student grades based on the combination of behavioral and traditional data.