Computational Methods for Designing Human-Centered Recommender Systems

Recommender Systems (RecSys) are ubiquitous and have become inseparable parts of our everyday lives in applications such as e- commerce, entertainment, and social media, providing personalized user experiences. Their impact is also growing in education, health- care, tourism, transport, and logistics, enhancing decision-making and user engagement. Hence, designing modern day RecSys requires a multi-disciplinary approach, incorporating machine learning, information retrieval, and human-computer interaction (HCI). This tutorial focuses on human-centric RecSys design, emphasizing both computational methods and user-centered principles. Participants will learn fundamental concepts, advanced algorithms, and practical implementation, with case studies linking visual arts and healthcare applications
Read More    Enroll Now
Explore

Computational Methods for Designing Human-Centered Recommender Systems: A Case Study Approach Intersecting Visual Arts and Healthcare

This tutorial approaches Recsys from a human-centred perspective, looking at the interface and algorithm studies that advance understanding of how system designs can be tailored to users’ objectives and needs while taking into account external factors such as commercialization. This course takes a case study approach to RecSys from an HCI perspective intersecting visual arts with healthcare application.

This tutorial is part of a Doctoral course on Recommender Systems that is being offered as part of PhD training in the Doctoral School of Computer Science and Computer Engineering at the University of Luxembourg.

Checkout our Blog to see featured student projects from previous editions of the course.

Program

Monday, October 14, 2024, from 9:00 - 12:30 at the 18th ACM Conference on Recommender Systems (RecSys 2024) Bari, Italy.

For detail about the venue and schedule updates, please check the official Recsy 2024 conference program

Register Now
Part 1

Introduction: Human-Centered RecSys

  • Personalisation: The Big Picture
  • Cyber-Physical-Social Systems (CPSS)
  • Why Human-Centered?
  • HC Recsys pipeline & a Framework for Modelling

    • Part 2 A

    HC RecSys Pipeline: A case-study approach

  • Personalized Visual Art Recommendation (VA RecSys)
  • Problem Formulation
  • Solutions: Computational methods
    • Data Representation learning (Unimodal & Multi-modal)
    • Transfer learning
    • Neural topic modelling
  • Evaluation Techniques & Results
    • Offline Experiments
    • User Studies
    • Online Experiments

    Part 2 B

    HC RecSys Pipeline: A case-study approach

  • Personalized Visual Art & Path Recommen- dation: A multi-stakeholder aware approach
  • Problem Formulation
  • Solutions: Computational methods
    • Mixed-Integer Linear Programming (MILP) Algorithms
  • Evaluation Techniques & Results
    • Offline Experiments
    • User Studies
    • Online Experiments

    Part 2 C

    VA RecSys for Post-Intensive Care Syndrome (PICS) intervention

  • Problem formulation
  • Solutions: Unimodal & multimodal VA Recsys
  • Ensuring safe and sensitive deployment
  • User evaluation & Results
  • Lessons learned

    • Part 3

    Hands-on

  • Practical exercises on the case studies

    • Contact

    Course Instructor

    Dr. Bereket YILMA

    University of Luxembourg, Luxembourg


    Course Instructor

    University of Luxembourg


    Maison du Nombre
    6, Avenue de la Fonte
    L-4364, Esch-sur-Alzette, Luxembourg


    bereket.yilma@uni.lu


    (+352) 466-644-9732

    Copyright © Bereket A. YILMA | All rights reserved