Collaborative Filtering Prone to Popularity Bias, Resulting in Overrepresentation of Popular Items in the Recommendation Outputs

March 1, 2022

Understanding the risks posed by popularity bias in collaborative filtering algorithms, leading to an overrepresentation of popular items within AI-generated recommendations. This awareness is crucial for ensuring responsible AI governance and the provision of diverse, trustworthy recommendation outputs. Stay tuned for updates on Project Cerebellum's efforts to establish guardrails for safe and secure AI practices.

For those interested in shaping the future of responsible AI and contributing to the HISPI Project Cerebellum TAIM (Govern, Map, Measure, Manage), JOIN US.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

MEASURE 2.6 — similarity 0.702, rank 1. TAIM detail and related incidents →
MAP 1.6 — similarity 0.685, rank 2. TAIM detail and related incidents →
MEASURE 1.3 — similarity 0.685, rank 3. TAIM detail and related incidents →

Alleged deployer: facebook, linkedin, youtube, twitter, netflix
Alleged developer: facebook, linkedin, youtube, twitter, netflix
Alleged harmed parties: facebook-users, linkedin-users, youtube-users, twitter-users, netflix-users

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/168

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.