Algorithmic Effects on the Diversity of Consumption on Spotify

Anderson, Ashton; Maystre, Lucas; Anderson, Ian; Mehrotra, Rishabh; Lalmas, Mounia

doi:10.1145/3366423.3380281

Cited by 137 publications

(103 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The role of recommendation devices in fostering diversity is assuredly at the heart of a fast-growing literature focused on a myriad of platform types, where music streaming is only one application case among many. Contrarily to popular assumptions about so-called "filter bubbles", the emerging empirical picture suggests that recommendation algorithms generally seem to increase diversity and serendipity [1,4,19,29,35,36], even though recent results on specific platforms such as Spotify or YouTube tend to suggest otherwise [3,37], while explicit personalization or "self-selection" also appear to induce algorithmic reinforcement and confinement, for instance regarding news consumption [14,51]. Most of this literature works at the aggregate level without distinguishing populations of users who may differently use or respond to algorithmic guidance.…”

Section: Related Workmentioning

confidence: 99%

Follow the guides: disentangling human and algorithmic curation in online music consumption

Villermet

Poiroux

Moussallam³

et al. 2021

Fifteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

The role of recommendation systems in the diversity of content consumption on platforms is a much-debated issue. The quantitative state of the art often overlooks the existence of individual attitudes toward guidance, and eventually of different categories of users in this regard. Focusing on the case of music streaming, we analyze the complete listening history of about 9k users over one year and demonstrate that there is no blanket answer to the intertwinement of recommendation use and consumption diversity: it depends on users. First we compute for each user the relative importance of different access modes within their listening history, introducing a trichotomy distinguishing so-called 'organic' use from algorithmic and editorial guidance. We thereby identify four categories of users.We then focus on two scales related to content diversity, both in terms of dispersion -how much users consume the same content repeatedly -and popularity -how popular is the content they consume. We show that the two types of recommendation offered by music platforms -algorithmic and editorial -may drive the consumption of more or less diverse content in opposite directions, depending also strongly on the type of users. Finally, we compare users' streaming histories with the music programming of a selection of popular French radio stations during the same period. While radio programs are usually more tilted toward repetition than users' listening histories, they often program more songs from less popular artists. On the whole, our results highlight the nontrivial effects of platform-mediated recommendation on consumption, and lead us to speak of 'filter niches' rather than 'filter bubbles'. They hint at further ramifications for the study and design of recommendation systems. CCS Concepts: • Human-centered computing → Empirical studies in HCI; • Information systems → Multimedia streaming; Recommender systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Follow the guides: disentangling human and algorithmic curation in online music consumption

Villermet

Poiroux

Moussallam³

et al. 2021

Fifteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

show abstract

“…With regard to education, Bolson, Rodrigues and Revello de Lima (2020) comment that these technologies allow not only to broaden the horizons of students within the classroom, instigating their curiosities, but also contribute positively to corporate environments and for the family environment; providing an abundance of information that is available in the most varied media and channels, favoring curiosity, arousing exploration and research, as well as the digital autonomy of these generations (Rolim, Mello & Costa, 2017;cf. Uline, 1996).…”

Section: Exponential and Disruptive Technologiesmentioning

confidence: 99%

Information management and exponencial learning in Society 5.0: foundations for an integrated system

Oliveira¹,

Pereira²

2021

Preprint

View full text Add to dashboard Cite

Humanity uses the information to reduce discomfort, effort, and errors. With this in mind, headphones were invented, so that the music could be heard without disturbing others; home appliances were created to reduce the effort in washing dishes or clothes, and calculating machines were developed so that human distraction would not cause an accident in calculating the weight that a bridge could support. But the information has hardly changed the certainties and doubts about which university course to take, what the individual's daily life will be like in the academic path, which includes when and what to study, what to eat and what to do in their free time. Is there a hobby that the individual could enjoy, but does not know? There is greater access to information, but many times instead of navigating this information, the individual sinks in the middle of it. In addition, teachers and students remain stuck in the curriculum, with the same menu being provided for carnivores, vegetarians or lactose intolerants. If before, students were beaten with wood in their hands and were attacked psychologically, having to learn the same as everyone else, nowadays only hands are no longer suffering from wood, at least in theory. Reducing mistakes in choosing a higher education course and the whole life that governs the academic path, can be a dream for many individuals, and software that collects enough information to learn about the individual and guide him/her on the beginning of their academic path could be an extremely useful compass. Knowing which university course to take, which discipline, among the electives, to choose, knowing the right time to get up according to the circadian cycle, what to do throughout the day, what and when to eat, are all informational outputs that utopian software could provide so as to contribute to the individual's success. But would such informational management drastically reduce the hassles, efforts, and mistakes to be made, or would it negatively affect the individual's life due to lack of experience with failure? Is the information learned from failure really necessary for success? This article starts from a brief history about information in education, analyses information management in Society 5.0, in the era of exponential technologies, and the construction of competence matrices. The goal is to start a debate about the delegation and trust in software, the management of personal information for individual success.

show abstract

“…We compute the embedding vector of a query as a weighted average of the Word2Vec [23] vectors of the clicked results. Similarly, the vector representation of a track is computed based on its co-occurrence statistics across playlists and the vector representation of an artist is the weighted average of the tracks they have performed [4].…”

Section: Standalone Featuresmentioning

confidence: 99%

Query Understanding for Surfacing Under-served Music Content

Tomasi¹,

Mehrotra²,

Pappu

et al. 2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

Self Cite

View full text Add to dashboard Cite

Platform ecosystems have witnessed an explosive growth by facilitating interactions between consumers and suppliers. Search systems powering such platforms play an important role in surfacing content in front of users. To maintain a healthy, sustainable platform, systems designers often need to explicitly consider exposing under-served content to users, content which might otherwise remain undiscovered. In this work, we consider the question when we might surface under-served content in search results, and investigate ways to provide exposure to certain content groups. We propose a framework to develop query understanding techniques to identify potential non-focused search queries on a music streaming platform, where users' information needs are non-specific enough to expose under-served content without severely impacting user satisfaction. We present insights from a search ranker deployed at scale and present results from live A/B test targeting a random sample of 72 million users and 593 million sessions, to compare performance of different methods considered to identify non-focused queries for surfacing under-served content.

show abstract

Algorithmic Effects on the Diversity of Consumption on Spotify

Cited by 137 publications

References 23 publications

Follow the guides: disentangling human and algorithmic curation in online music consumption

Follow the guides: disentangling human and algorithmic curation in online music consumption

Information management and exponencial learning in Society 5.0: foundations for an integrated system

Query Understanding for Surfacing Under-served Music Content

Contact Info

Product

Resources

About