Projects per year
Abstract
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.
Original language | English |
---|---|
Title of host publication | WWW'22: Proceedings of the ACM Web Conference 2022 |
Number of pages | 10 |
Publisher | Association for Computing Machinery |
Publication date | Apr 2022 |
Pages | 2719–2728 |
ISBN (Electronic) | 9781450390965 |
DOIs | |
Publication status | Published - Apr 2022 |
MoE publication type | A4 Article in conference proceedings |
Event | The ACM Web Conference - Lyon, France Duration: 25 Apr 2022 → 29 Apr 2022 |
Fields of Science
- 113 Computer and information sciences
- recommender systems
- random walks
- radicalization
- polarization
- extremist content
- filter bubbles
Projects
- 1 Active
-
MLDB: Model Management Systems: Machine learning meets Database Systems
Gionis, A., Mathioudakis, M., Merchant, A., Pai, S. G., Svana, M. & Wang, Y.
Suomen Akatemia Projektilaskutus
01/09/2019 → 31/12/2023
Project: Academy of Finland: Academy Project
Prizes
-
Best paper award at the ACM Web Conference 2022
Fabbri, Francesco (Recipient), Wang, Yanhao (Recipient), Bonchi, Francesco (Recipient), Castillo, Carlos (Recipient) & Mathioudakis, Michael (Recipient), 29 Apr 2022
Prize: Prizes and awards