What's Hot? Estimating Country-specific Artist Popularity

Predicting artists that are popular in certain regions of the world is a well desired task, especially for the music industry. Also the cosmopolitan and cultural-aware music aficionado is likely be interested in which music is currently “hot” in other parts of the world. We therefore propose four approaches to determine artist popularity rankings on the country-level. To this end, we mine the following data sources: page counts from Web search engines, user posts on Twitter, shared folders on the Gnutella file sharing network, and playcount data from last.fm. We propose methods to derive artist rankings based on these four sources and perform cross-comparison of the resulting rankings via overlap scores. We further elaborate on the advantages and disadvantages of all approaches as they yield interestingly diverse results.