Technology
Predicting MOOC Dropout with Machine Learning
Quick fact
Studies show that in many MOOCs, more than 90% of enrolled learners never finish the course, yet machine learning models can identify at-risk students with up to 80-90% accuracy using only the first week of clickstream data.
Why this is interesting
You've signed up for a free online course, but after a week, you stop watching the videos. Could an algorithm have seen it coming?