Beyond the mainstream music industry, new alternative practices and platforms for sharing and engaging with music have emerged. These are used especially by new and independent artists, although some established musicians are also using them as secondary mechanisms for the ad hoc circulation of their music, and they are beginning to have a significant effect on the music industry as they provide a mechanism for independent artists to attract fans while remaining outside of conventional industry processes.
This project examines engagement practices around one leading site, SoundCloud (which attracts 175 million unique monthly listeners), using big data from SoundCloud and social media. SoundCloud users are able to listen to tracks provided by artists and attach their own comments to them; they are also able to share links to these tracks through social media such as Facebook and Twitter. We draw on the Application Programming Interfaces (APIs) of these platforms to capture a large dataset of SoundCloud comments as well as tweets and Facebook posts sharing links to SoundCloud, and use a combination of automated content analysis of these comments and posts, and network analysis of the overlap in user populations between individual songs, to measure the affinity between songs and develop a framework for making automatic song recommendations to users as they engage with SoundCloud content.
Funding / Grants
- UA-DAAD Research Collaboration (2017 - 2018)