Generative AI for Synthetic Data in Learning Analytics

Opportunities, Methods, and Challenges

A hands-on workshop exploring the intersection of generative AI and synthetic data generation for privacy-preserving learning analytics research.

April 28, 2026 - LAK26 Conference

Part of LAK26 Pre-Conference Workshops

Our workshop is officially part of the LAK26 pre-conference schedule. View all workshops and register through the official conference portal.

View LAK26 Official Schedule

Tuesday, April 28, 2026

Morning Half-Day Session

9:00 AM - 12:30 PM

3.5 Hours Workshop

In-Person Event

LAK26 Conference Venue

About the Workshop

Learning analytics research is often constrained by limited access to authentic educational data due to ethical, legal, and privacy concerns. Synthetic data has emerged as a promising approach to address these challenges by generating artificial datasets that preserve the statistical properties and utility of real data while protecting individual privacy.

The recent rise of Generative AI (GenAI), including large language models (LLMs), diffusion models, and generative adversarial networks (GANs), has significantly advanced the quality and scalability of synthetic data generation.

Tabular Data

Generate synthetic educational datasets using CTGAN and advanced techniques

Text Generation

Create synthetic assignments, feedback, and dialogues using LLMs

Visual Content

Generate educational images and visualizations with diffusion models

Workshop Schedule

9:00 - 9:15

Opening & Introductions

Welcome, participant introductions, and workshop overview

9:15 - 9:45

Invited Keynote Talk

Foundational perspectives on synthetic data in learning analytics

9:45 - 10:30

Lightning Talks

Short presentations from participants and Q&A sessions

10:30 - 10:45

Break

Networking and refreshments

10:45 - 11:30

Hands-on Tutorial: Tabular Data

CTGAN, SDV frameworks, and evaluation metrics

11:30 - 12:15

Hands-on Tutorial: Text & Multimodal

LLMs for text generation and diffusion models for images

12:15 - 12:30

Discussion & Wrap-up

Community roadmap synthesis and closing remarks

Workshop Organizers

Farhad Vadiee

Farhad Vadiee

Researcher

Centre for the Science of Learning & Technology (SLATE)
University of Bergen, Norway

Qinyi Liu

Qinyi Liu

PhD Candidate

Centre for the Science of Learning & Technology (SLATE)
University of Bergen, Norway

Pengfei Li

Pengfei Li

PhD Candidate

Centre for the Science of Learning & Technology (SLATE)
University of Bergen, Norway

Oscar Deho

Oscar Deho

Postdoctoral Research Fellow

Centre for Change and Complexity in Learning (C3L)
University of South Australia, Australia

Mohammad Khalil

Mohammad Khalil

Associate Professor

Centre for the Science of Learning \& Technology (SLATE)
University of Bergen, Norway

Srecko Joksimovic

Srecko Joksimovic

Associate Professor

Centre for Change and Complexity in Learning (C3L)
University of South Australia, Australia

Datasets & Resources

Datasets Coming Soon!

We are preparing synthetic educational datasets that will be made available to workshop participants. These datasets will demonstrate various synthetic data generation techniques across different educational contexts.

Stay tuned for announcements about dataset availability and access instructions.

Tools & Frameworks

CTGAN, SDV, OpenAI APIs, Diffusion Models

Reading Materials

Curated papers and resources on synthetic data

Code Examples

Jupyter notebooks and implementation guides

Registration & Contact

Registration

Registration for this workshop is handled through the LAK26 official registration process. Please visit the LAK26 conference website for registration details and deadlines.

Get in Touch

For any questions about the workshop, lightning talk submissions, or to discuss potential topics, please contact the organizers:

Email the Organizers

Send your inquiries to any of our team members listed above

Propose a Lightning Talk

Share your research, use cases, or position papers

Ask Questions

Technical questions, tool requirements, or participation details

Lightning Talk Topics

  • Use cases of synthetic data in learning analytics
  • Novel methods for generating or evaluating synthetic data
  • Ethical, legal, or methodological reflections
  • Position papers on opportunities and challenges

Organizing Institutions

SLATE
University of Bergen, Norway
C3L
University of South Australia