DESI ONLINE SEMINAR
Abstract
YouTube has revolutionized the way people discover and consume videos, becoming one of the primary news sources for Internet users. Since content on YouTube is generated by its users, the platform is particularly vulnerable to misinformative and conspiratorial videos. Even worse, the role played by YouTube’s recommendation algorithm in unwittingly promoting questionable content is not well understood, and could potentially make the problem even worse. This can have dire real-world consequences, especially when pseudoscientific content is promoted to users at critical times, e.g., during the COVID-19 pandemic.
In this work, we set out to characterize and detect pseudoscientific misinformation on YouTube. We collect 6.6K videos related to COVID-19, the flat earth theory, the anti-vaccination, and anti-mask movements; using crowdsourcing, we annotate them as pseudoscience, legitimate science, or irrelevant. We then train a deep learning classifier to detect pseudoscientific videos with an accuracy of 76.1%. Next, we quantify user exposure to this content on various parts of the platform (i.e., a user’s homepage, recommended videos while watching a specific video, or search results) and how this exposure changes based on the user’s watch history. We find that YouTube’s recommendation algorithm is more aggressive in suggesting pseudoscientific content when users are searching for specific topics, while these recommendations are less common on a user’s homepage or when actively watching pseudoscientific videos. Finally, we shed light on how a user’s watch history substantially affects the type of recommended videos.
Bio
I am an Assistant Professor of Computer Engineering and Informatics.
I hold a PhD from Duke University since 2010. My research interests include trust-aware design of distributed systems, device-centric authentication, federated identity management, discrimination based on personal data, cybersafety (cyberbullying detection, cybergrooming detection, characterization and detection of hate speech, detection of inappropriate videos targeting young children, and characterization and suppression of false information), transactional workload scalability, measurement of blockchain systems.
I have published articles in the most influential conferences and journals of Networked Systems, including ACM SIGCOMM, USENIX NSDI, ACM IMC, USENIX ATC, AAAI ICWSM, IEEE INFOCOM, IEEE ICDCS, IEEE/ACM Transactions on Networking, and ACM Transactions on the Web.
I have extensive experience leading EU-funded projects. Specifically, I was the technical manager of the ReCRED project (Horizon 2020 Innovation Action - 2014) and the coordinator of the ENCASE project (Horizon 2020 Marie Curie RISE - 2015).
Our work on fringe web communities, hate speech, disinformation and disturbing content on YouTube had extensive coverage in major news outlets, including The New York Times, Washington Post, The Atlantic, New Scientist, Business Insider, Quartz, Wired, and El Pais.
I am also the co-director of the Network Systems and Science Research Laboratory and a member of the Board of Directors of the Research Centre of Excellence on Interactive media, Smart systems and Emerging technologies (RISE).
Join Zoom Meeting
https://unil.zoom.us/j/95089763957
Meeting ID: 950 8976 3957
Join by SIP
95089763957@zoomcrc.com