Last Updated on September 18, 2023 by Janina Wildermuth

We are proud to announce that researchers at MediaFutures will be presenting a total of five papers at RecSys ’23.

The 17th ACM Conference on Recommender Systems (RecSys ’23) is the largest community gathering on recommender systems research in the world and will take place in Singapore from September 18th to 22nd, 2023. Over 1000 people are invited to discuss and present their latest research in the field.

Among them, our MediaFutures researchers have the honor of presenting their latest work on novel recommendation systems for news, food and movies.

These are the papers which will be presented.

1. Evaluating The Effects of Calibrated Popularity Bias Mitigation: A Field Study. Klimashevskaia, A., Elahi, M., Jannach, D., Sjærven, L., Tessem, A. & Trattner, C. ACM RecSys 2023 LBR, 2023.
2. The Interplay between Food Knowledge, Nudges, and Preference Elicitation Methods Determines the Evaluation of a Recipe Recommender System. El Majjodi, A., Starke, A., Elahi, M. & Trattner, C., INTRS workshop at ACM RecSys 2023, 2023.

4. Understanding How News Recommender Systems Influence Selective Exposure. Seddik, K., Knudsen, E., Trilling, D. & Trattner, C. BehavRec workshop at ACM RecSys 2023, 2023.

5. Using Visual and Linguistic Framing to Support Sustainable Decisions in an Online Store. Starke, A.D., Emami, K., Makarova, A., Ferwerda, B., IntRS’23 workshop at ACM RecSys 2023, 2023. 

BehavRec ’23: Recommendations for Behavior Change

MediaFutures director Christoph Trattner will also be co-organising the workshop BehavRec at the conference on September 19th from 9:00 to 10:30.

The workshop centers on persuasive recommender systems, providing valuable insights into creating recommendations that encourage positive changes in behavior. The agenda encompasses a wide array of subjects, including behavior change recommender systems in domains such as health, food, education, and beyond.

Attendees will have the opportunity to explore user interfaces, such as visual, context-aware, and conversational interfaces. Additionally, innovative methodologies such as controllability, transparency, and context-awareness in recommendation design will be examined. Ethical and privacy considerations within the realm of behavior change technology will be addressed, alongside discussions on behavior change theories and persuasion techniques.

NORMalize: Normative Design and Evaluation of Recommender Systems

MediaFutures Key Researcher Alain Starke co-organizes a workshop called ‘NORMalize’, on normative design and evaluation of recommender systems at RecSys ’23.

NORMalize sets out to provide a platform for researchers and practicioners from different domains to discuss challenges related to the normative design and evaluation of recommender systems.

In this workshop they examine the question whether news recommenders should provide sufficiently diverse recommendations. Should we strive to expose users to content that contradicts their political beliefs?

It is RecSys’ first Workshop on Normative Design and Evaluation of Recommender Systems and will take place on tuesday, 19th of september both online and at the conference in Singapore.