Zhang, E. Riedl, V. Petrushin, S. Pal & J. Spoelstra; JMLR W&CP 18:215–229,
Committee Based Prediction System for Recommendation: KDD Cup 2011, Track2
This paper describes a solution to the 2011 KDD Cup competition, Track2:
discriminating between highly rated tracks and unrated tracks in a Yahoo! Music dataset. Our
approach was to use supervised learning based on 65 features generated using various techniques
such as collaborative ﬁltering, SVD, and similarity scoring. During our modeling stage, we
created a number of predictors including logistic regression, artiﬁcial neural networks and
gradient-boosted decision trees. To further improve robustness and reduce the variance, we used
three of our top performing models and took a weighted average for the ﬁnal submission, which
achieved 4.3768% error.
Page last modified on Tue May 29 10:23:35 2012.