Our solution relies on Knowledge Tracing Machines:
@inproceedings{Vie2019,
Author = {{Vie}, Jill-J{\^e}nn and {Kashima}, Hisashi},
Booktitle = {Proceedings of the 33th {AAAI} Conference on Artificial Intelligence},
Title = {{Knowledge Tracing Machines: Factorization Machines for Knowledge Tracing}},
Pages = {750--757},
Url = {http://arxiv.org/abs/1811.03388},
Year = 2019}
To be precise, at the time of the competition our solution relied on Deep Factorization Machines, but in the meantime Knowledge Tracing Machines proved to be less prone to overfitting.
- Install the requirements (e.g. in a virtualenv)
- Download the datasets from SLAM 2018 in the current folder