Introduction to Data Analysis Competition: Rating Prediction
Overview
§ The goal is to improve your model’s performance as much as possible.
§ You can improve your model’s performance through model selection and hyperparamet er tuning.
§ You can submit your model’s results to Kaggle and view your scores in real-time on the le aderboards.
• You can only submit a maximum of 20 per day.
Rules
§ In this competition, you will use Surprise, the python library for recommender systems.
§ You must participate as an individual team in this competition.
Evaluation
Submissions are evaluated on the Root Mean Square Error (RMSE) between the ground truth rating and your predicted rating.
• Please refer to the link.
Dataset
Movie-Rating Prediction Dataset
• The dataset consists of three columns.
user_id, item_id, rating
• You should train your model with this interaction data.
Competition
§ You will enter the competition individually through Kaggle.
§ Kaggle Site: https://www.kaggle.com
§ Competition URL: https://www.kaggle.com/t/7c309521f9294bc4b5d42cabfdb32f31
Competition
Kaggle-site
① Overview: Description of the task
② Data: Files to be used in the competition
③ Leaderboard: Current score updated in real time
④ Rules: What you must follow in the competition
⑤ My submissions: Files you submitted
⑥ Submit Predictions: Submission
