=Paper=
{{Paper
|id=Vol-3667/GenAILA-Preface
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-3667/GenAILA-Preface.pdf
|volume=Vol-3667
}}
==None==
https://ceur-ws.org/Vol-3667/GenAILA-Preface.pdf
Generative AI for Learning Analytics (GenAI-LA):
Exploring Practical Tools and Methodologies
Lixiang Yan1, Andy Nguyen2, Lele Sha1, Jionghao Lin3, Mutlu Cukurova4, Kshitij Sharma5,
Roberto Martinez-Maldonado1, Linxuan Zhao1, Yuheng Li1, Yueqiao Jin1, and Dragan
Gašević1
1 Monash University, Wellington Rd, Clayton VIC, 3800, Australia
2 University of Oulu, Pentti Kaiteran katu 1 Linnanmaa, Finland
3 Carnegie Mellon University, 4804 Forbes Avenue, Pittsburgh, PA 15213, USA
4 University College London, 20 Bedford Way, London, WC1H 0AL, United Kingdom
5 Norwegian University of Science and Technology, IT-bygget, 147, Gløshaugen, Norway
Abstract
The GenAI-LA workshop aims to highlight the significant impact generative artificial intelligence
(GenAI) can have on learning analytics (LA). GenAI offers a wide range of benefits, including the ability
to process unstructured data automatically, create adaptive educational materials, and improve the way
LA findings are communicated through engaging stories and thorough explanations. This workshop
aims to bring together a diverse mix of professionals, including learning scientists, LA practitioners,
software developers, and AI experts, to engage in meaningful discussions and collaborations. Our goal
is to thoroughly explore and envision how GenAI can play a crucial role in advancing both the research
and application of learning analytics. We received a total of 12 paper submissions. Following a thorough
peer review process, we accepted seven full papers and two short papers. Each paper presents a unique
perspective on the potential of GenAI to transform the field of LA and education, from addressing data
sparsity in intelligent tutoring systems to supporting self-regulated learning and beyond.
Keywords
Generative artificial intelligence, learning analytics, educational technologies, large language models
Preface
In this collection of workshop papers, we delve into the innovative intersection of generative
artificial intelligence (GenAI) and learning analytics, exploring a wide array of applications and
methodologies aimed at enhancing educational experiences and outcomes. Each paper presents
a unique perspective on the potential of GenAI to transform the field of education, from
addressing data sparsity in intelligent tutoring systems to supporting self-regulated learning and
beyond. Here is a brief introduction to each paper:
3DG: A Framework for Using Generative AI for Handling Sparse Learner Performance Data
From Intelligent Tutoring Systems by Liang Zhang et al. introduces the 3DG framework, a novel
approach that combines tensor factorization with advanced generative models to address the
challenge of sparse data in intelligent tutoring systems. This paper showcases how the framework
can generate scalable, personalised simulations of learning performance, highlighting the
potential of GenAI in creating more effective educational technologies.
An AI Agent Facilitating Student Help-Seeking: Producing Data on Student Support Needs
by Joonas Merikko and Anni Silvola examines the use of large language models (LLMs) to facilitate
students' help-seeking behaviours. The authors discuss the development of a support bot
prototype that leverages GPT-4 and WhatsApp APIs, aiming to lower barriers to help-seeking and
gather data on student support needs.
Enhancing Trust in Generative AI: Investigating Explainability of LLMs to Analyse
Confusion in MOOC Discussions by Yuanyuan Hu et al. explores the potential of explainable AI
(XAI) methods to enhance trust in GenAI tools. Through a pilot study, the paper demonstrates
how XAI can identify indicators of confusion in MOOC discussions, advocating for the integration
of XAI in GenAI applications to improve learning analytics solutions.
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
Generative AI for Critical Analysis: Practical Tools, Cognitive Offloading and Human
Agency by Simon Buckingham Shum discusses the implications of GenAI for critical analysis
tasks. The paper highlights practical examples where GenAI can support intellectual engagement
and examines the balance between the benefits and risks of cognitive offloading to AI.
Generative Multimodal Analysis (GMA) for Learning Process Data Analytics by Ridwan
Whitehead et al. introduces the GMA method, a systematic framework for applying GenAI to the
analysis of multimodal learning process data. This paper presents a case study to demonstrate
the method's applicability and effectiveness, contributing to the advancement of learning
analytics methods.
GPT-3.5, GPT-4, Bard, and Claude’s Performance on the Chinese Reading Comprehension
Test by Bor-Chen Kuo et al. assesses the performance of advanced GenAI models in Chinese
reading comprehension tasks. The study compares these models' performances with that of fifth-
grade students, offering insights into the educational applications of GenAI.
Supporting Self-Regulated Learning with Generative AI: A Case of Two Empirical Studies
by Jacqueline Wong and Olga Viberg explores the use of GenAI chatbots to enhance self-regulated
learning and learning performance. The paper presents preliminary results from two empirical
studies, contributing to the discussion on personalising SRL support using AI.
Supporting Student Decisions on Learning Recommendations: An LLM-Based Chatbot with
Knowledge Graph Contextualization for Conversational Explainability and Mentoring by
Hasan Abu-Rasheed et al. proposes an LLM-based chatbot that utilises a knowledge graph to
support students in understanding learning-path recommendations. The paper discusses the
chatbot's development and evaluation, highlighting its potential in conversational explainability.
Tamil Co-Writer: Towards Inclusive Use of Generative AI for Writing Support by Antonette
Shibani et al. extends the application of GenAI to the development of a writing aid prototype for
the Tamil language. The paper emphasises the importance of inclusive AI tools that support
linguistically diverse learners, showcasing the potential of GenAI to enhance writing skills and
productivity.
Together, these papers represent a comprehensive exploration of the transformative potential of
GenAI in learning analytics and education, offering valuable insights and innovative solutions to
the challenges faced by learners, educators, and researchers in the field.