New distribution channels like music streaming platforms paved way for making more and more diverse music available to users. Thus, music recommender systems got in the focus of research in academia as well as industry. Collaborative filtering-based recommender systems have been proven useful, but there is space left for improvements by adapting this general approach to better fit to the music recommendations problem. In this work, we incorporate context-based information about the music consumption into the recommendation process. This information is extracted from playlist names, which are analyzed and aggregated into so-called "contextual clusters". We find that the listening context plays an important role and thus allows for providing recommendations reaching precision values 33% higher than traditional approaches. Hence, the main contribution of this paper is a new method that extracts and integrates contextual information from playlist names into the recommendation process for improving music recommendations.
展开▼
