Applying Data Mining Techniques to Identify Success Factors in Students Enrolled in Distance Learning: A Case Study

Distance learning is now a key component in higher level education. Given the high dropout rates and the important investments in distance learning it is of utmost concern to determine the most critical data in the success and failure of students. In this article we data mine enrollment profiles, educational background and students' data from the Open University System and Distance Learning of the National Autonomous University of Mexico to determine the key factors that drive success and failure, creating a relevant predictive model using a Naive Bayes classifier. We have found that the number of subjects approved and their average qualification in the first semester are part of the most interesting predictors of student success.