We investigate ensemble learning methods for hybrid music recommender algorithms, combining a social and a contentbased recommender algorithm as weak learners by applying a combination rule to unify the weak learners' output. A first experiment suggests that such a combination can already reduce the mean absolute prediction error compared to the weak learners' individual errors.

This article presents an approach to automatically create virtual communities of users with similar music preferences in a distributed system. Our goal is to create personalized music channels for these communities using the content shared by its members in peer-to-peer networks for each community. To extract these communities a complex network theoretic approach is chosen. A fully connected graph of users is created using epidemic protocols. We show that the created graph sufficiently converges to a graph created with a centralized algorithm after a small number of protocol iterations. To find suitable techniques for creating user communities, we analyze graphs created from real-world recommender datasets and identify specific properties of these datasets. Based on these properties, different graph-based community-extraction techniques are chosen and evaluated. We select a technique that exploits identified properties to create clusters of music listeners. The suitability of this technique is validated using a music dataset and two large movie datasets. On a graph of 6,040 peers, the selected technique assigns at least 85% of the peers to optimal communities, and obtains a mean classification error of less than 0.05% over the remaining peers that are not assigned to the best community.

Recommender systems are increasingly being employed to personalize services, such as on the web, but also in electronics devices, such as personal video recorders. These recommenders learn a user profile, based on rating feedback from the user on, e.g., books, songs, or TV programs, and use machine learning techniques to infer the ratings of new items.The techniques commonly used are collaborative filtering and naive Bayesian classification, and they are known to have several problems, in particular the cold-start problem and its slow adaptivity to changing user preferences. These problems can be mitigated by allowing the user to set up or manipulate his profile.In this paper, we propose an extension to the naive Bayesian classifier that enhances user control. We do this by maintaining and flexibly integrating two profiles for a user, one learned by rating feedback, and one created by the user. We in particular show how the cold-start problem is mitigated.