Tutorials
How to Build Explainable Recommender Systems using Path Reasoning on Knowledge Graphs: A Tutorial with hopwise
Presenter: Ludovico Boratto, Gianni Fenu, Francesca Maridina Malloci, Mirko Marras, Giacomo Medda and Alessandro Soccol
Description:
Recommender systems are a core testbed for user modeling and personalization research in UMAP. As these systems increasingly rely on complex models, the need for transparency, user trust, and regulatory compliance has made explainability a core requirement. Path reasoning over knowledge graphs offers a viable foundation for generating human-understandable recommendations, yet its adoption remains limited due to fragmented tools and high coding complexity. This tutorial introduces hopwise, a unified and extensible framework for building explainable-by-design recommender systems through path reasoning on knowledge graphs. In line with UMAP principles, the tutorial treats knowledge graphs and learning over them as a form of user modeling, where structured representations of users, items, and their relations drive both personalization and explanation. Through a progressive, hands-on workflow, participants will move from standard recommendation datasets to knowledge-graph representations, run and compare reinforcement learning–based and language model–based path reasoning methods, generate natural-language explanations, and evaluate recommendation utility and explanation quality. The tutorial also guides participants on how to extend the framework to new domains, reasoning or evaluation perspectives, enabling direct reuse.
Interface-Aware Recommender Systems
Presenters: Santiago de Leon-Martinez, Behnam Rahdari, Robert Moro, Peter Brusilovsky and Maria Bielikova
Link: https://umap26.kinit.sk/
Description:
Despite the wide-spread use of multi-list or carousel (Netflix-like) interfaces in e-commerce and streaming services, there is little academic research (less than 30 papers), especially when compared to works for single-list interfaces. Recent eye tracking results have shown that users browse multi-list and carousels significantly differently than other interfaces. Carousels are much more complex, allowing a wide-range of browsing/interaction sequences with multiple topic defined-lists that can be swiped to see more items. To account for this complexity and improve recommendations, recommender systems should be designed specifically for the interfaces they are used on, in other words interface-aware recommenders.
This tutorial introduces researchers and industry practitioners to the growing area of interface-aware recommenders. The tutorial provides an introduction to varying interfaces and their impact on user behavior, an overview of the research and insights for improving interface specific systems, the open problems and challenges that have not been addressed, and tools/datasets to help tackle these problems. The tutorial’s goal is to provide a strong basis and tools that participant can use to build improved user-centric systems that are interface-aware and also encourage future research in this area.
From Modeling and Profiling to Auditing: the Evolving Role of the User in Recommender System Research
Presenters: Lorenzo Porcaro and Erasmo Purificato
Description:
The proposed tutorial is designed to provide the UMAP community with a structured exploration of the transition from traditional user modeling to modern algorithmic auditing, emphasizing the user’s active role in ensuring the accountability of recommender systems. First, we offer a thorough retrospective on the established fields of user profiling and modeling, covering both their historical origins and the technical milestones that have shaped how user preferences are captured and exploited by personalized systems. We then analyze the evolving conceptual framework and terminology of the field to resolve ambiguities regarding the interpretation of key notions. As the core of our tutorial, we detail the recent shifts that move beyond passive profiling toward participatory auditing frameworks, where users actively engage in surfacing biases and systemic harms. In particular, we examine and discuss advancements in the following areas: the evolution toward pseudo-explicit profiling, universal user representations, and diverse methods of user-led auditing. Attendees will be encouraged to participate in interactive discussions and Q&A sessions throughout the program to ensure an engaging learning experience.
Inferential and Causal Principles for Better Understanding Interactive Systems
Presenters: David Rohde
Description: The practice of building high performance personalized interactive systems is founded on an artful combination of diverse machine learning methodologies including collaborative filtering, content based recommendation, contextual bandits, off policy estimation, click models, attribution and A/B testing. While combining these methods has been spectacularly successful, each methods finds justification using its own stylized protocols and there is little attention to the over-aching principles behind building reward optimizing recommender systems. This tutorial will focus on inferential principles including causal inference and relate these principles to current best practice in machine learning. The inferential principles covered will include Bayesian decision theory, coherence, the likelihood and conditionality principle as well as causal principles such as ignoreability, the do-calculus, Rubin Causal model, randomization and A/B testing. These foundations will then be used to investigate the differences between recommender systems best practices and approaches directly informed by inferential and causal principles, this section will both challenge best practice and the applicability of academic approaches. A lot of attention will be given to the pervasive problem of self confounding which will cover how engineering and machine learning best practice often results in production systems that unnecessarily suffer from confounding.