CARS - Workshop on Context-Aware Recommender Systems
Contextual information has been widely recognized as an important modeling dimension in social science and technological disciplines, and it is becoming more and more important for enhancing the relevancy of the recommendation results. Several context-aware recommender systems (CARS) workshops have been organized in the past, where the addition of contextual information to traditional recommender systems has been discussed. While a substantial amount of research has already been performed, many existing approaches to CARS focus on the so-called `representational view' that incorporates pre-defined and static contextual factors (such as time and location) in the recommendation process.
However, in the past few years, new CARS techniques have been introduced, such as sequence-aware recommender systems and latent context-aware recommender systems. Moreover, inferring implicit contexts in real-time (online) environments and measuring business metrics for multiple new application areas, such as education, health, cooperative work and affective computing, require the modeling of complex, partially observable and dynamic contextual factors.
The CARS 2025 workshop focuses on rethinking the role of contextual information in modern recommender systems, particularly in light of their growing adoption in mobile, IoT, and multi-stakeholder environments. These emerging domains demand novel context modeling strategies that are not only adaptive and dynamic but also address critical challenges such as privacy, security, fairness, and bias mitigation. As the contextual landscape becomes more complex and sensitive, the next generation of CARS should adopt explainable, flexible, and privacy-preserving techniques, such as federated learning and secure computation, to safeguard user data without compromising recommendation quality. The workshop aims to explore these challenges and opportunities by fostering cross-disciplinary discussions, identifying key research questions, and advancing innovative approaches for building trustworthy and effective context-aware recommender systems.
CARS 2025 workshop will be held in person on September 22th, 2025 in the morning, between 9 am - 12:45 pm, and will be co-located with RecSys'25 in Prague.
