Let's Influence Algorithms Together: How Millions of Fans Build Collective Understanding of Algorithms and Organize Coordinated Algorithmic Actions

Research Background and Issues

• Identified Problems or Challenges:

  • Previous studies primarily focus on how users understand and interact with algorithms on an individual level, with limited research on how communities collectively comprehend algorithms and organize collective actions against them.
  • There is currently a phenomenon of large-scale fan-driven collective algorithmic actions, which have profound impacts on platforms and society, such as challenging the integrity of platform algorithms, distorting algorithmic outputs, and triggering public trust crises in platforms.
  • Such collective actions demonstrate the potential of user groups to actively reshape algorithms to achieve specific goals, but they also pose challenges to platform governance and fairness.

• Significance of the Study:

  • Understanding these large-scale fan-driven collective actions is crucial for improving platform governance, enhancing transparency, and resisting algorithmic abuse.
  • It also provides a new perspective for discussing algorithmic accountability and ethical issues.

• Research Motivation and Related Work:

  • Existing user-driven algorithm research mostly explores individual-level "algorithmic folk theories" and small-scale algorithmic resistance behaviors.
  • This study extends this perspective by investigating how fan communities engage in large-scale, cross-platform, and cross-cultural collective actions and the organizational mechanisms involved.

Proposed Solution

• Proposed Approach:

  • This study, conducted over two years of ethnographic research (2021-2023), explores how fan communities form collective algorithmic understandings, mobilize millions of users, and organize cross-platform fan algorithmic actions.
  • The authors analyzed the understanding and participation processes of 43 core fans (whose leadership is critical) and 15 ordinary fans in algorithmic actions.

• Innovations:

  • For the first time, this study expands algorithmic folk theories from an individual perspective to a collective level, revealing how theoretical concepts are translated into practical actions through community collaboration.
  • It demonstrates how fan communities overcome the "collective action dilemma" of large-scale participation, such as persuading ordinary fans to invest time and effort and providing easy-to-understand participation instructions.
  • By combining emotional mobilization and technical strategies, it offers a new paradigm for analyzing group digital interventions.

• Implementation Steps and Key Techniques:

  1. Core fans decode platform algorithms through repeated experiments and observations, forming folk understandings as the strategic foundation.
  2. Emotional promotional strategies (e.g., "emotional persuasion" or "guilt-based mobilization") are used to motivate ordinary fans to participate.
  3. Simple and user-friendly "algorithm tutorials" are designed to lower participation barriers (e.g., through visual guides or small tools), guiding ordinary users in large-scale data operations (such as liking, commenting, and sharing).
  4. Manipulation strategies are adjusted based on the algorithmic characteristics of different countries and platforms, such as using different repeat viewing or comment ranking strategies for YouTube and Weibo.

Research Findings

• Specific Findings:

  • The study reveals the process by which fans form multi-platform algorithmic understandings through folk theories and test and adjust these understandings in actions.
  • It systematically analyzes the mobilization strategies of core fans, including how they address the skepticism of ordinary fans (e.g., by simplifying operational instructions) and maintain participant loyalty through "emotional narratives."
  • It preliminarily explores how fan communities use semi-automated tools (e.g., Android automation clickers) to reduce operational complexity.

• Advantages Over Existing Solutions:

  • Compared to individual-level algorithmic studies, this research provides a new perspective on large-scale digital collective actions, showcasing the complete transformation process from understanding to practice.
  • By introducing a dual driving mechanism of emotion and technology, it offers theoretical support for how large groups can organize effective computer-supported collective actions.

• Experimental or Evaluation Results:

  • Core fans reported successful cases of operations, such as optimizing actions to push specific topics onto Weibo's trending list.
  • Fans successfully adapted to the opacity of different platform algorithms and iteratively developed effective action models through experience accumulation.

• Limitations and Future Directions:

  • The study focuses only on fan communities, and its applicability to other user groups requires further exploration.
  • Most analyses in the study focus on successful cases, lacking detailed observations of factors contributing to collective action failures.
  • Future research could expand to other non-fan communities' collective behaviors or explore the ethics and governance strategies of collective actions.

Through this study, the authors not only deepen the understanding of algorithmic folk theories but also provide important practical recommendations for platform design and algorithmic transparency. These insights have broad academic and practical implications, particularly in balancing user agency and platform norms within the complex ecosystem of social media.