In this repository, we focus on the recommendation task on a subset of the famous Netflix competition. The selected dataset contains 10, 000 users and 10, 000 movies (items).
- First, we prepare the data in the desired form in data.py.
- Then we implement a user-based collaborative filtering (CF) algorithm and several variants in CF.py.
- At last, we implement the matrix factorization (MF) algorithm in MF.py.
We conduct all experiments in main.py. We also provide a brief reports to introduce our implementation details.
To run our codes, first unzip the data.zip file and then change to the code directory and use the following command.
python main.py