SXSW EDU 2020
Artificial Intelligence for Course Recommendation
With over 800 courses in a single platform, it becomes a challenge to decide what to study next. For a single student, topics are somewhat similar while their interests are diverse. Wasting time is a big worry.
Traditional recommender algorithms failed miserably in this case study, but recurrent neural networks, presented statistically significant results with students being more engaged, for a longer period of time and, most importantly, finishing more courses.
Share this idea
- Understand how recommenders can help students
- Understand how to tackle that challenge with recurrent neural networks and how to measure its impact with students
- Interpret the results found in the case study, avoiding some biases while doing so
- Guilherme Silveira, Head of Education, Alura
Guilherme Silveira, Head Of Education, Alura