The second project of the EPFL Machine Learning course, which aims to construct a Recommender System with good predictive performance using various Machine Learning Algorithms. The given report explains our machine learning procedure in detail together with the well documented python code in the scripts folder.
-
crowdAI competition link: https://www.crowdai.org/challenges/epfl-ml-recommender-system
-
Group name: THREE COMMA CLUB
-
Leaderboard (https://www.crowdai.org/challenges/epfl-ml-recommender-system/leaderboards)
- RMSE = 1.017, which is the 2nd best predictive performance. (username: efeacer)
scripts
: contains python code that established our machine learning procedurereport.pdf
: project report explaining our machine learning procedure in .pdf formatproject2_description.pdf
: assignment description given by EPFL
Please don't hesitate to contact the authors about any questions about the project or machine learning in general:
- Efe Acer: efeacer@gmail.com
- Murat Topak: murattopak01@gmail.com
- Daniil Dmitriev: dmitrievdaniil97@gmail.com
Licensed under the MIT License
© 2018 Efe Acer