Algorithms Recommendations on Social Media and Political Positions, with Tim Faverjon

How recommendation algorithms operate on social media to establish relationships with the political positions of users? In order to answer this question, Tim Faverjon designed various models and analysed their predictions, specifically focusing on political attitudes and socio-demographic characteristics. He emphasises the importance of looking inside the algorithms rather than just observing their outcomes to understand their influence on users.

Tim Faverjon, PhD candidate at the médialab, data science engineer and mathematician, carries out his research at the interface between machine learning and sociology. His current research focuses on recommendation algorithms and politics: what do algorithms “know” about user ideology? How is this information used? What impact on the digital public debate?

Hosted by Ausha. See ausha.co/privacy-policy for more information.