We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Retrieval

Title: Allowing for equal opportunities for artists in music recommendation

Abstract: Promoting diversity in the music sector is widely discussed on the media. While the major problem may lie deep in our society, music information retrieval contributes to promoting diversity or may create unequal opportunities for artists. For example, considering the known problem of popularity bias in music recommendation, it is important to investigate whether the short head of popular music artists and the long tail of less popular ones show similar patterns of diversity---in terms of, for example, age, gender, or ethnic origin---or the popularity bias amplifies a positive or negative effect. I advocate for reasonable opportunities for artists---for (currently) popular artists and artists in the long-tail alike---in music recommender systems. In this work, I represent the position that we need to develop a deep understanding of the biases and inequalities because it is the essential basis to design approaches for music recommendation that provide reasonable opportunities. Thus, research needs to investigate the various reasons that hinder equal opportunity and diversity in music recommendation.
Comments: 3 pages, position paper, 1st Workshop on Designing Human-Centric MIR Systems, Delft, 2019
Subjects: Information Retrieval (cs.IR); Computers and Society (cs.CY)
Cite as: arXiv:1911.05395 [cs.IR]
  (or arXiv:1911.05395v1 [cs.IR] for this version)

Submission history

From: Christine Bauer [view email]
[v1] Wed, 13 Nov 2019 10:58:30 GMT (187kb)

Link back to: arXiv, form interface, contact.