Collaborative filtering ensemble for ranking
Michael Jahrer, Andreas Töscher · Knowledge Discovery and Data Mining · 2011
This paper provides the solution of the team commendo on the Track2 dataset of the KDD Cup 2011 Dror et al.. Yahoo Labs provides a snapshot of their music-rating database as dataset for the competition, consisting of approximately 62 million ratings from 250k users on 300k items. The dataset includes hierachical information about the items. The goal of the competition is to distinguish beteen and Not items of a user. The rating scale is discrete and ranges from 0 to 100, while a High rating is a rating≥80. The error measure is the percent of false rated tracks over all users, known as the fractions of misclassifications. The task is to minimize this error rate, hence the ranking should be optimized. Our final submission is a blend of different collaborative filtering algorithms enhanced, with basic statistics. The algorithms are trained consecutively and they are blended together with a neural network. Each of the algorithms optimizes a rank error measure.