Music Recommendation System for Million Song Dataset Challenge

摘要

In this paper a system that took 8th place in Million Song Dataset challengeis described. Given full listening history for 1 million of users and half oflistening history for 110000 users participatints should predict the missinghalf. The system proposed here uses memory-based collaborative filteringapproach and user-based similarity. MAP@500 score of 0.15037 was achieved.