# Rewiring What-to-Watch-Next Recommendations to Reduce Radicalization Pathways

arXiv: https://arxiv.org/abs/2202.00640 · HTML: https://mlsanity.com/paper/2202.00640/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 60 references that are in the map.
- First posted on arXiv 1 Feb 2022.

## Builds on (most cited first, 21 of 21)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Fake News Detection on Social Media: A Data Mining Perspective](https://mlsanity.com/paper/1708.01967/) (2017)
- [Equity of Attention: Amortizing Individual Fairness in Rankings](https://mlsanity.com/paper/1805.01788/) (2018)
- [The Rise of Social Bots](https://mlsanity.com/paper/1407.5225/) (2014)
- [Fairness in Rankings and Recommendations: An Overview](https://mlsanity.com/paper/2104.05994/) (2021)
- [Auditing Radicalization Pathways on YouTube](https://mlsanity.com/paper/1908.08313/) (2019)
- [Detecting Hate Speech in Social Media](https://mlsanity.com/paper/1712.06427/) (2017)
- [Quantifying Controversy in Social Media](https://mlsanity.com/paper/1507.05224/) (2015)
- [NELA-GT-2018: A Large Multi-Labelled News Dataset for The Study of Misinformation in News Articles](https://mlsanity.com/paper/1904.01546/) (2019)
- [Recommendations and User Agency: The Reachability of Collaboratively-Filtered Information](https://mlsanity.com/paper/1912.10068/) (2019)
- [The Effect of People Recommenders on Echo Chambers and Polarization](https://mlsanity.com/paper/2112.00626/) (2021)
- [Minimizing Polarization and Disagreement in Social Networks](https://mlsanity.com/paper/1712.09948/) (2017)
- [Disturbed YouTube for Kids: Characterizing and Detecting Inappropriate Videos Targeting Young Children](https://mlsanity.com/paper/1901.07046/) (2019)
- [Algorithmic Extremism: Examining YouTube's Rabbit Hole of Radicalization](https://mlsanity.com/paper/1912.11211/) (2019)
- [Exposure Inequality in People Recommender Systems: The Long-Term Effects](https://mlsanity.com/paper/2112.08237/) (2021)
- [Approximation Algorithms for Reducing the Spectral Radius to Control Epidemic Spread](https://mlsanity.com/paper/1501.06614/) (2015)
- [Quantifying Availability and Discovery in Recommender Systems via Stochastic Reachability](https://mlsanity.com/paper/2107.00833/) (2021)
- [RePBubLik: Reducing the Polarized Bubble Radius with Link Insertions](https://mlsanity.com/paper/2101.04751/) (2021)
- [Absorbing random-walk centrality: Theory and algorithms](https://mlsanity.com/paper/1509.02533/) (2015)
- [Improving the betweenness centrality of a node by adding links](https://mlsanity.com/paper/1702.05284/) (2017)
- [Coverage Centrality Maximization in Undirected Networks](https://mlsanity.com/paper/1811.04331/) (2018)

## Built on by (most cited first, 6 of 6)

- [Cascade-based Echo Chamber Detection](https://mlsanity.com/paper/2208.04620/) (2022)
- [Harm Mitigation in Recommender Systems under User Preference Dynamics](https://mlsanity.com/paper/2406.09882/) (2024)
- [Manifestations of Xenophobia in AI Systems](https://mlsanity.com/paper/2212.07877/) (2022)
- [Refined Causal Graph Structure Learning via Curvature for Brain Disease Classification](https://mlsanity.com/paper/2506.15708/) (2025)
- [Reducing Exposure to Harmful Content via Graph Rewiring](https://mlsanity.com/paper/2306.07930/) (2023)
- [Efficient Edge Rewiring Strategies for Enhancing PageRank Fairness](https://mlsanity.com/paper/2602.02512/) (2026)

## Abstract (arXiv)

Recommender systems typically suggest to users content similar to what they consumed in the past. If a user happens to be exposed to strongly polarized content, she might subsequently receive recommendations which may steer her towards more and more radicalized content, eventually being trapped in what we call a "radicalization pathway". In this paper, we study the problem of mitigating radicalization pathways using a graph-based approach. Specifically, we model the set of recommendations of a "what-to-watch-next" recommender as a d-regular directed graph where nodes correspond to content items, links to recommendations, and paths to possible user sessions. We measure the "segregation" score of a node representing radicalized content as the expected length of a random walk from that node to any node representing non-radicalized content. High segregation scores are associated to larger chances to get users trapped in radicalization pathways. Hence, we define the problem of reducing the prevalence of radicalization pathways by selecting a small number of edges to "rewire", so to minimize the maximum of segregation scores among all radicalized nodes, while maintaining the relevance of the recommendations. We prove that the problem of finding the optimal set of recommendations to rewire is NP-hard and NP-hard to approximate within any factor. Therefore, we turn our attention to heuristics, and propose an efficient yet effective greedy algorithm based on the absorbing random walk theory. Our experiments on real-world datasets in the context of video and news recommendations confirm the effectiveness of our proposal.