MORS 2021: 1st Workshop On Multi-Objective Recommender Systems.pdf

Recommender systems are software tools used in various domains to support users in finding relevant items, products, and services. Historically, the main criterion for a successful recommender system was the relevance of the recommended items to the user. However, real-world recommender systems often take into account multiple objectives, which can be from the users' perspective or from other stakeholders such as item providers and those impacted by the recommendations. For example, in restaurant recommendations, factors like users' taste, diet restrictions, proximity, and price should be considered. Similarly, in education, a system should balance student preferences with utility for learning. Objectives may also come from stakeholders like platform owners or society. The MORS workshop addressed the challenges of producing recommendations in multi-objective and multi-stakeholder settings, including topics like value-aware recommendation, trade-off between relevance and bias, and fairness-aware recommender systems. The workshop aimed to understand various objectives and goals for recommender systems, algorithms to generate recommendations in a multi-objective environment, and new evaluation approaches. The workshop organizers were Himan Abdollahpouri, Mehdi Elahi, Masoud Mansoury, Shaghayegh Sahebi, Zahra Nazari, Allison Chaney, and Babak Loni. The program committee consisted of academic and industry researchers who reviewed submissions. The workshop followed a peer review process and had a tentative timeline from April to September 2021. The workshop program included a keynote by Shankar Kalyanaraman titled "Measuring and mitigating long-term effects of recommender systems: A framework and a call to action.