Algorithmic Audit of Personalisation Drift in Polarising Topics on TikTok

• URL: http://arxiv.org/abs/2603.20723v1
• Date: Sat, 21 Mar 2026 09:13:06 GMT
• Title: Algorithmic Audit of Personalisation Drift in Polarising Topics on TikTok
• Abstract summary: We investigate the extent to which TikTok steers content toward specific topics and polarities over time.<n>Specifically, we investigated a preference-aligned drift, a polarisation-topic drift and a polarisation-stance drift.<n>Overall, our findings provide evidence that recommendation trajectories differ markedly across topics.
• Score: 6.386379303219094
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Social media platforms have become an integral part of everyday life, serving as a primary source of news and information for many users. These platforms increasingly rely on personalised recommendation systems that shape what users see and engage with. While these systems are optimised for engagement, concerns have emerged that they may also drive users toward more polarised perspectives, particularly in contested domains such as politics, climate change, vaccines, and conspiracy theories. In this paper, we present an algorithmic audit of personalisation drift on TikTok in these polarising topics. Using controlled accounts designed to simulate users with interests aligned with or opposed to different polarising topics, we systematically measure the extent to which TikTok steers content exposure toward specific topics and polarities over time. Specifically, we investigated: 1) a preference-aligned drift (showing a strong personalisation towards user interests), 2) a polarisation-topic drift (showing a strong neutralising effect for misinformation-themed topics, and a high preference and reinforcement of interest of US politic topic); and 3) a polarisation-stance drift (showing a preference of oppose stance towards US politics topic and a general reinforcement of users' stance by recommending items aligned with their stance towards polarising topics). Overall, our findings provide evidence that recommendation trajectories differ markedly across topics, with some pathways amplifying polarised viewpoints more strongly than others and offer insights for platform governance, transparency and user awareness.

Related papers

• Tree of Preferences for Diversified Recommendation [54.183647833064136]
  We study diversified recommendation from a data-bias perspective.<n>Inspired by the outstanding performance of large language models (LLMs) in zero-shot inference leveraging world knowledge, we propose a novel approach.
  arXiv  Detail & Related papers  (2025-12-24T04:13:17Z)
• Constructing Political Coordinates: Aggregating Over the Opposition for Diverse News Recommendation [1.1787037402510556]
  News recommender systems (NRSs) have shown to be useful in minimizing political disengagement and information overload.<n>NRSs often conflate user interest with the partisan bias of the articles in their reading history.<n>Over extended interaction, this can result in the formation of filter bubbles and the polarization of user partisanship.
  arXiv  Detail & Related papers  (2025-11-14T23:04:04Z)
• Reading Between the Lines: A Study of Thematic Bias in Book Recommender Systems [0.0]
  This paper introduces and investigates thematic bias in book recommendations, defined as a disproportionate favouring or neglect of certain book themes.<n>We adopt a multi-stage bias evaluation framework to evaluate thematic bias in recommendations and its impact on different user groups.<n>Our findings show that thematic bias originates from content imbalances and is amplified by user engagement patterns.
  arXiv  Detail & Related papers  (2025-08-21T15:20:39Z)
• Personalized Topic Selection Model for Topic-Grounded Dialogue [24.74527189182273]
  Current models tend to predict user-uninteresting and contextually irrelevant topics. We propose a textbfPersonalized topic stextbfElection model for textbfTopic-grounded textbfDialogue, named textbfPETD. Our proposed method can generate engaging and diverse responses, outperforming state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-06-04T06:09:49Z)
• A First Look at Selection Bias in Preference Elicitation for Recommendation [64.44255178199846]
  We study the effect of selection bias in preference elicitation on the resulting recommendations. A big hurdle is the lack of any publicly available dataset that has preference elicitation interactions. We propose a simulation of a topic-based preference elicitation process.
  arXiv  Detail & Related papers  (2024-05-01T14:56:56Z)
• The Impact of Recommendation Systems on Opinion Dynamics: Microscopic versus Macroscopic Effects [1.4180331276028664]
  We study the impact of recommendation systems on users, both from a microscopic (i.e., at the level of individual users) and a macroscopic perspective. Our analysis reveals that shifts in the opinions of individual users do not always align with shifts in the opinion distribution of the population.
  arXiv  Detail & Related papers  (2023-09-16T11:44:51Z)
• Unpacking polarization: Antagonism and Alignment in Signed Networks of Online Interaction [0.3581083356941628]
  In the 20th century, major fault lines were formed by structural conflicts, like owners vs workers, center vs periphery, etc. We present the FAULTANA pipeline, a computational method to uncover major fault lines in data of signed online interactions. Our method makes it possible to quantify the degree of antagonism prevalent in different online debates, as well as how aligned each debate is to the major fault line.
  arXiv  Detail & Related papers  (2023-07-13T05:57:48Z)
• Unveiling the Hidden Agenda: Biases in News Reporting and Consumption [59.55900146668931]
  We build a six-year dataset on the Italian vaccine debate and adopt a Bayesian latent space model to identify narrative and selection biases. We found a nonlinear relationship between biases and engagement, with higher engagement for extreme positions. Analysis of news consumption on Twitter reveals common audiences among news outlets with similar ideological positions.
  arXiv  Detail & Related papers  (2023-01-14T18:58:42Z)
• Two-Stage Neural Contextual Bandits for Personalised News Recommendation [50.3750507789989]
  Existing personalised news recommendation methods focus on exploiting user interests and ignores exploration in recommendation. We build on contextual bandits recommendation strategies which naturally address the exploitation-exploration trade-off. We use deep learning representations for users and news, and generalise the neural upper confidence bound (UCB) policies to generalised additive UCB and bilinear UCB.
  arXiv  Detail & Related papers  (2022-06-26T12:07:56Z)
• Joint Multisided Exposure Fairness for Recommendation [76.75990595228666]
  This paper formalizes a family of exposure fairness metrics that model the problem jointly from the perspective of both the consumers and producers. Specifically, we consider group attributes for both types of stakeholders to identify and mitigate fairness concerns that go beyond individual users and items towards more systemic biases in recommendation.
  arXiv  Detail & Related papers  (2022-04-29T19:13:23Z)
• Perspective-taking to Reduce Affective Polarization on Social Media [11.379010432760241]
  We deploy a randomized field experiment through a browser extension to 1,611 participants on Twitter. We find that simply exposing participants to "outgroup" feeds enhances engagement, but not an understanding of why others hold their political views. framing the experience in familiar, empathic terms by prompting participants to recall a disagreement does not affect engagement, but does increase their ability to understand opposing views.
  arXiv  Detail & Related papers  (2021-10-11T20:25:10Z)
• News consumption and social media regulations policy [70.31753171707005]
  We analyze two social media that enforced opposite moderation methods, Twitter and Gab, to assess the interplay between news consumption and content regulation. Our results show that the presence of moderation pursued by Twitter produces a significant reduction of questionable content. The lack of clear regulation on Gab results in the tendency of the user to engage with both types of content, showing a slight preference for the questionable ones which may account for a dissing/endorsement behavior.
  arXiv  Detail & Related papers  (2021-06-07T19:26:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.