MT Serendipitous Recommender Systems by Balancing User Control and Diversity

Recommender systems recommend movies, restaurants or other items to an active user based on information about users and items in various application domains. Recommendation algorithms are often black-box systems, in that users cannot influence the recommendation process. However, it is desirable to provide users with more detailed control over recommendation strategies, and this is also required by the EU Digital Services Act (DSA). Influencing recommender systems seems more important in complex and high-risk domains such as real estate, travel and tourism, and high-end eCommerce. On the other hand, letting users influence the recommendation algorithm and resulting content may lead to filter bubbles. Diversity of recommendations is thus also desirable, and it is important to consider not just the user’s perspective, but other stakeholders’ as well. For example in tourism, popular destinations suffer from over-tourism, while other destinations would welcome more exposure by recommendation and related systems.

The main goal of Master's Thesis in Informatics (or related study programs) is to investigate this trade-off between user control and diversity in more detail. Determining the exact scenario and research setup is part of the task, but the focus should be on questions of whether users would accept more unexpected and „serendipitous“ items. For example, in the mentioned tourism domain, an idea could be to recommend alternative options to well-known and very popular options, that still fit the users’ preferences but are more diverse. Convincing users of recommended items may be helped by explanations and should be evaluated in a user study.

Prerequisites are high motivation and possibly some background in the topic area, e.g. experiments with users. Please send your application (brief CV, transcript of records, and short motivation statement including ReactJS experience) to Wolfgang Wörndl (woerndl@cit.tum.de) by March 20th, 2026, with a response and decision soon after. The actual start of the thesis project is flexible.