The RecSys Conference 2024

During the Doctoral Symposium, Brett Binst will present his work from the Serendipity Engine project.

Recommender systems can assist in various user tasks and serve diverse values, including exploring the item space. Serendipity has recently received considerable attention, often seen as a way to broaden users’ tastes and counteract filter bubbles. However, the field of research on serendipity is fragmented regarding its evaluation methods, which impedes the progress of knowledge accumulation. This research plan proposes two studies to address these issues. First, a systematic literature review will be conducted to provide insights into how serendipity is currently studied in the field. This review will serve as a reference for novice researchers and help mitigate fragmentation by presenting a thorough overview of the field. This systematic literature review has already revealed a significant gap: the lack of a validated, widely accepted method for evaluating serendipity. Therefore, the second part of this research plan is to develop a validated questionnaire, the serendiptionnaire, to measure serendipity. This tool will provide a ground truth for evaluating serendipity, aiding in answering fundamental questions within the field and validating offline metrics.

During the poster session on Tuesday, Ulysse Maes will present joint work with Lien Michiels and Annelien Smets.

Previous research has used Large Language Models (LLMs) to develop personalized Conversational Recommender Systems (CRS) with text-based user interfaces (UIs). However, the potential of LLMs to generate interactive graphical elements that enhance user experience remains largely unexplored. To address this gap, we introduce “GenUI(ne) CRS,” a novel framework designed to leverage LLMs for adaptive and interactive UIs. Our framework supports domain-specific graphical elements such as buttons and cards, in addition to text-based inputs. It also addresses the common LLM issue of outdated knowledge, known as the “knowledge cut-off,” by implementing Retrieval-Augmented Generation (RAG). To illustrate its potential, we developed a prototype movie CRS. This work demonstrates the feasibility of LLM-powered interactive UIs and paves the way for future CRS research, including user experience validation, transparent explanations, and addressing LLM biases.

During the poster session on Thursday, Hanne Vandenbroucke will present joint work with Annelien Smets.

Recommender systems are increasingly used by news media organizations. Existing literature examines various aspects of news recommender systems (NRS) from a computational, user-centric, or normative perspective. Yet research advocates studying the complexities of real-world applications around NRS. Recently, a multi-stakeholder approach to NRS has been adopted, allowing to understand different stakeholder perspectives on NRS development and evaluation within the news organization. However, little research has been done on the different key performance indicators (KPIs) and metrics considered valuable by different stakeholders. Based on 11 interviews with professionals from two commercial news publishers, this paper demonstrates that stakeholders prioritize distinct KPIs and metrics related to the reach-engagement-conversion-retention funnel. The evaluation of NRS performance is often limited to short-term metrics like CTR, overlooking the multiplicity of stakeholders involved. Our findings reveal how different purposes, KPIs, and metrics are valued from the journalistic, commercial, and tech logic. In doing so, this paper contributes to the multi-stakeholder approach to NRS, advancing our understanding of the real-world complexity of NRS development and evaluation.

During the MuRS: Music Recommender Systems Workshop 2024, Rebecca Salganik will present joint work with Valdy Wiratama, Heritiana Ranaivoson and Adelaida Afilipoaie (FairMuse).

The proliferation of digital technologies in the distribution of digital content has prompted concerns about the effects on cultural diversity in the digital era. The concept of discoverability has been presented as a theoretical tool through which to consider the likelihood that content will be interacted with. The multifaceted nature of this broad theme has been explored through a variety of domains that explore the ripple effects of platformization, each with its own unique lexicography. However, there is yet to be a unified framework through which to consider the complex pathways of discovery. In this work we present the discovery ecosystem, consisting of six individual, interconnected components, that encompass the pathway of discovery from start to finish.

On October 18, Lien Michiels is co-organizing the Second Workshop on Normative Design and Evaluation of Recommender Systems.