Explaining User Models with Diferent Levels of Detail for Transparent Recommendation: A User Study Mouadh Guesmi University of Duisburg-Essen Duisburg, Germany mouadh.guesmi@stud.uni.de Mohamed Amine Chatti University of Duisburg-Essen Duisburg, Germany mohamed.chatti@uni-due.de Laura Vorgerd University of Duisburg-Essen Duisburg, Germany laura.vorgerd@stud.uni-due.de Thao Ngo University of Duisburg-Essen Duisburg, Germany thao.ngo@uni-due.de Shoeb Joarder University of Duisburg-Essen Duisburg, Germany shoeb.joarder@uni-due.de Qurat Ul Ain University of Duisburg-Essen Duisburg, Germany qurat.ain@stud.uni.de Arham Muslim National University of Sciences and Technology Islamabad, Pakistan arham.muslim@seecs.edu.pk ABSTRACT In this paper, we shed light on explaining user models for trans- parent recommendation while considering user personal character- istics. To this end, we developed a transparent Recommendation and Interest Modeling Application (RIMA) that provides interactive, layered explanations of the user model with three levels of detail (basic, intermediate, advanced) to meet the demands of diferent types of end-users. We conducted a within-subject study (N=31) to investigate the relationship between personal characteristics and the explanation level of detail, and the efects of these two variables on the perception of the explainable recommender system with regard to diferent explanation goals. Based on the study results, we provided some suggestions to support the efective design of user model explanations for transparent recommendation. CCS CONCEPTS · Human-centered computing → Interactive systems and tools; · Computing methodologies → Artifcial intelligence. KEYWORDS intelligent explanation interfaces; recommender systems; explain- able recommendation; explainable user modeling, personal charac- teristics ACM Reference Format: Mouadh Guesmi, Mohamed Amine Chatti, Laura Vorgerd, Thao Ngo, Shoeb Joarder, Qurat Ul Ain, and Arham Muslim. 2022. Explaining User Models Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for proft or commercial advantage and that copies bear this notice and the full citation on the frst page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specifc permission and/or a fee. Request permissions from permissions@acm.org. UMAP ’22 Adjunct, July 4ś7, 2022, Barcelona, Spain © 2022 Association for Computing Machinery. ACM ISBN 978-1-4503-9232-7/22/07. . . $15.00 https://doi.org/10.1145/3511047.3537685 with Diferent Levels of Detail for Transparent Recommendation: A User Study. In Adjunct Proceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization (UMAP ’22 Adjunct), July 4ś7, 2022, Barcelona, Spain. ACM, New York, NY, USA, 9 pages. https://doi.org/10.1145/3511047. 3537685 1 INTRODUCTION Recommender systems (RS) are one of many adaptive systems that leverage user models to deliver relevant content to their end-users. User models have been enriched with various features such as open- ness, scrutability, and explainability. These features are the most investigated ones by researchers in view of their signifcant impact on the user’s perception of adaptive systems and their outcomes [7, 21]. Opening the user model means allowing users to see how the system is perceiving them in a human-understandable form, which will lead to several benefts such as improving the accuracy of the model [12]. Scrutinizing the user model is a concept built on top of openness and is related to user control in a sense that, in addition to letting the users inspect their models, they can interact with them (e.g., edit the content, provide more information) [12]. Explaining the user model consists of providing explanations about how these models were generated [19]. Recently, research on ex- plainable recommendations started to focus on explaining the user models (i.e., explaining the recommendation input) as an alterna- tive to revealing the inner working of the system (i.e., explaining the recommendation process) or justifying the recommended items (i.e., explaining the recommendation output) [8, 21]. In addition to the explanation scope (i.e., input, process, output), another crucial design choice in explainable recommendation re- lates to the level of explanation detail that should be provided to the end-user [2]. Users may not be interested in all the information that the explanation can produce [38]. Diferent users have diferent needs for explanation and explanations may cause negative efects (e.g., high cognitive load, confusion, lack of trust) if they are difcult to understand [18, 27, 30, 52, 53]. The majority of current designs 175