QUARE @ Recsys 2023
2nd Workshop on Measuring the Quality of Explanations
in Recommender Systems (virtual)
quā-rē, adv. [quae-res] I. Interrog., by what means? how? II. From what cause, on what account, wherefore, why.
What is QUARE?
QUARE—Measuring Quality of Explanations in Recommender Systems—is the second workshop that focuses on evaluation methodologies for explanations in recommender systems. During this edition, we bring together researchers and practitioners from academia and industry to facilitate discussions about the main issues and best practices in the respective areas, identify possible synergies, and outline priorities regarding future research directions. Additionally, we want to stimulate reflections around methods to systematically and holistically assess explanation approaches, impact, and goals, at the interplay between organisational and human values. To that end, this workshop aims to co-create a research agenda for evaluating the quality of explanations for recommender systems.
This year, the workshop is co-located with the 17th ACM Conference on Recommender Systems and will be organized as a virtual event, on the 11th of September, 2023.
Focus
During the first edition of the QUARE workshop, co-located with SIGIR 2022, we identified the need for measuring the quality of explanations for recommender systems from multiple perspectives, including:
1. user-centric explanation evaluation;
2. design requirements for explanations from actual recommender systems target users;
3. diverse, interdisciplinary, and multi-stakeholder evaluation.
In planning the current edition, we build on the lessons learned from the first edition. Accordingly, this workshop's edition aims to shape a research agenda for evaluating the quality of explanations for recommender systems. We aim for a discussion-centric event, where the main outcome would be a document outlining this research agenda. We plan to engage a diverse, multidisciplinary community of academic and industry researchers and practitioners to facilitate the exchange of perspectives and solutions to the challenges and issues around the evaluation of the quality of explanations for recommender systems. The organisers, having diverse backgrounds, will facilitate discussions on three threads, which correspond to the main topics of the workshop:
Generating explanations: How to balance user needs and organisational objectives when explaining recommendations?
Evaluating explanations: How to evaluate the quality of explanations through both subjective and objective measures?
Operationalising explanations: What could be a typical MVP (Minimum Viable Product) for explanations of recommendations? What are the most basic features we may need?
All the presentations and the interactive discussions will happen online as the workshop is virtual.
Read more about the first edition of the QUARE workshop in our technical report:
Alessandro Piscopo, Oana Inel, Sanne Vrijenhoek, Martijn Millecamp, and Krisztian Balog. 2022. Report on the 1st Workshop on Measuring the Quality of Explanations in Recommender Systems (QUARE 2022) at SIGIR 2022. SIGIR Forum 56, 2, Article 11 (dec 2022), 16 pages, https://dl.acm.org/doi/pdf/10.1145/3582900.3582915.