MT Increasing Privacy in Group Recommender Systems through Round-by-Round Disclosure of Preferences

Recommender systems recommend movies, restaurants or other items to an active user or user group based on information about users and items. Submitting user preferences and personal information to services raises privacy issues. But this is also a concern within a group in group recommender systems, as users may not be comfortable sharing their whole profile with all group members.

The goal of the Master's Thesis project is to investigate selective and progressive user profile disclosure in group recommender systems from the users' perspective. The specific scenario can be chosen, for example, movies or point-of-interests recommendation for an ad-hoc group, i.e. the potential group members do not know each other (well). The idea is that users disclose selected information, such as ratings for genres or items, in a roundwise fashion. So all group members submit first some preferences, then recommendations are generated and inspected. Then users can submit additional preferences to receive better recommendations in subsequent rounds. One aspect is to investigate if users are willing to not only have their preferences used in the recommendation algorithm, but also explicitly shown to other group members, or under what conditions. Another issue is group dynamics, for example, some users may want to provide as much of their profile as possible, to dominate the recommendation process.

The course of action is as follows:

* Overview of relevant state-of-the-art in the research literature
* Designing the solution and implementing the interactive prototype 
* Conducting an evaluation with users using an appropriate methodology

The prototype application can be developed using any platform, for example, Web-based. Prerequisites are high motivation and good enough programming skills in the selected programming environment. Please send your application (brief CV, transcript of records and short motivation statement) to Wolfgang Wörndl (woerndl@in.tum.de) until September 28th, 2025 at the latest, the actual start of the thesis project is flexible.