Calibrated-Game-Recommender-

Calibrated Steam Game Recommender based on RecSys Paper "Calibrated Recommenders" by Harald Steck

Dependencies

To run this project you will need:

 - Numpy
 - Matplotlib
 - Pandas
 - Implicit

Treat data

Everything related to treat the data is on clean_data folder, you can use your own, or our small preprocessed dataset To use a larger dataset, put all json files from jmcauley steam dataset on the folder "data" and run "steam_games.py" and "user_item_playtime.py"

How to run

python3 recommender.py

How to use

You can change user, tweak settings using variables on code header

References:

Calibrated Recommendations Paper from Harald Stack: paper

Calibrated Recommendations Presentations: presentation

Steam dataset by Julian McAuley from University of California San Diego:
dataset