Crowdsourced Audit of Twitter’s Recommender Systems - Institut Curie Access content directly
Journal Articles Scientific Reports Year : 2023

Crowdsourced Audit of Twitter’s Recommender Systems

Audit crowdsourcé du système de recommandation de Twitter

Abstract

This research conducts an audit of Twitter's recommender system, aiming to examine the disparities between users' curated timelines and their subscription choices. Through the combined use of a browser extension and data collection via the Twitter API, our investigation reveals a high amplification of friends from the same community, a preference for amplifying emotionally charged and toxic tweets and an uneven algorithmic amplification across friends' political leaning. This audit emphasizes the importance of transparency, and increased awareness regarding the impact of algorithmic curation.
Fichier principal
Vignette du fichier
scientific_reports_twitter_bias_main.pdf (221.59 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04036232 , version 1 (19-03-2023)
hal-04036232 , version 2 (24-03-2023)
hal-04036232 , version 3 (03-07-2023)
hal-04036232 , version 4 (01-10-2023)

Identifiers

Cite

Paul Bouchaud, David Chavalarias, Maziyar Panahi. Crowdsourced Audit of Twitter’s Recommender Systems. Scientific Reports, 2023, 13 (1), ⟨10.1038/s41598-023-43980-4⟩. ⟨hal-04036232v4⟩
604 View
223 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More