Let's demolish my thesis: Is using Machine Learning to predict student failure a good idea?
Life at the Faculty of Engineering of the National University of Asunción is tough. With around 12.000 class inscriptions every semester, only around 6.000 end up with a passing grade. This, combined with the institutional weakness that characterizes Paraguayan state institutions and regular heat waves (and regular power cuts), creates a very tense environment for students, teachers and authorities alike.
As I was choosing a research topic to finish off my time in university as an engineering student at this very university, I was offered the opportunity to work on a prediction, binary classification and ranking system using Data Science and Machine Learning. What exactly am I predicting? Student failure.
In this session, we will explore how this system is designed and discuss who it really benefits: the students, the faculty, or the system itself.
|Discussion - Capped