Gave a keynote on “Responsible Recommendation in the Age of Generative AI” at the ROEGEN workshop at #RecSys2024 back in October; since I've obtained the recording, I now have an edited transcript available: https://md.ekstrandom.net/talks/2024/roegen-transcript
(will add a link to the video if and when that's public.)
Responsible Recommendation in the Age of Generative AI Talk Transcript
Transcript for “Responsible Recommendation in the Age of Generative AI”, my keynote talk for the Workshop on Risks, Opportunities, and Evaluation of Generative Models in Recommender Systems (ROEGEN) at RecSys 2024.
Me, recovering from
#recsys2024 and reading posts from people I met last week about them traveling to (or already being at)
#ecai2024 and
#cikm2024For those interested… slides and bibliography for my talk at ROEGEN at
#RecSys2024 are here:
https://md.ekstrandom.net/talks/2024/roegenResponsible Recommendation in the Age of Generative AI
Keynote talk at ROEGEN workshop at RecSys 2024.
How easy is it to get into a #FilterBubble on #SocialMedia, and how can we analyze them? 🤨
Our master student Luka developed the tool SOAP that allows creating and analyzing filter bubbles in personalized social media feeds.
He'll will present the tool and our findings on Friday at "NORMalize 2024: The Second Workshop on the Normative Design and Evaluation of Recommender Systems", co-located with ACM #RecSys2024!
📄 https://www.alexandria.unisg.ch/handle/20.500.14171/120987
💾 https://github.com/LukaBekavac/SOAP/
@InteractionsUniSG
From Walls to Windows: Creating Transparency to Understand Filter Bubbles in Social Media
Social media platforms play a significant role in shaping public opinion and societal norms. Understanding this influence requires examining the diversity of content that users are exposed to. However, studying filter bubbles in social media recommender systems has proven challenging, despite extensive research in this area. In this work, we introduce SOAP (System for Observing and Analyzing Posts), a novel system designed to collect and analyze very large online platforms (VLOPs) data to study filter bubbles at scale. Our methodology aligns with established definitions and frameworks, allowing us to comprehensively explore and log filter bubbles data. From an input prompt referring to a topic, our system is capable of creating and navigating filter bubbles using a multimodal LLM. We demonstrate SOAP by creating three distinct filter bubbles in the feed of social media users, revealing a significant decline in topic diversity as fast as in 60min of scrolling. Furthermore, we validate the LLM analysis of posts through an inter-and intra-reliability testing. Finally, we open source SOAP as a robust tool for facilitating further empirical studies on filter bubbles in social media.
Really glad to see
@Riedl highlighting the need to reflect on positionality in AI work!
#RecSys2024First of the three
#recsys2024 keynotes:
@Riedl on Human-centered explainable AI
#xaiKicking off
#recsys2024 with a keynote from
@Riedl on human-centered explainable AI.
In 5 minutes we have
@jvinagre’s keynote on “Recommender systems research and the EU’s Digital Services Act and AI Act”. Don’t miss it
#RecSys2024!
Really gonna miss the lively
#recsys2024 hashtag on the site formerly known as Twitter 😔
Good morning
#RecSys2024! We've gotten started with our opening session — it isn't too late to come join us!
https://facctrec.github.io/facctrec2024/program/Program
A venue for discussing problems of social responsibility in recommendation
Michael Ekstrand
Alan Said
Jannis Strecker-Bischoff
FAccTRec Workshop