Improving Quality of Service for Radio Station Hosting: An Online Recommender System Based on Information Fusion

Abstract

We present a new recommender system developed for the Russian interactive radio network FMhost. The system aims to improve the quality of this service; it is designed specifically to deal with small datasets, overcoming the shortage of data on observed user behavior. The underlying model combines a collaborative user-based approach with information from tags of listened tracks in order to match user and radio station profiles. It follows an adaptive online learning strategy based on both user history and implicit feedback. We compare the proposed algorithms with industry standard methods based on Singular Value Decomposition (SVD) in terms of precision, recall, and Normalized Discounted Cumulative Gain (NDCG) measures; experiments show that in our case the fusion-based approach produces the best results.