Leveraging Hierarchical Knowledge Structures for Adaptive and Emotionally Engaging E-Learning - MSc Thesis Defense by: Y (Kitty) Duong

Tuesday, May 6, 2025 - 14:00

The School of Computer Science is pleased to present…

Leveraging Hierarchical Knowledge Structures for Adaptive and Emotionally Engaging E-Learning
MSc Thesis Defense by: Y (Kitty) Duong

 

Date: Tuesday, May 6th, 2025

Time: 2:00 PM

Location: Dillon Hall, Room 255

 

Abstract:

The COVID-19 pandemic has accelerated the evolution of E-learning systems, offering high-quality educational experiences comparable to traditional classrooms from virtually anywhere. However, this rapid growth has also presented a critical challenge, particularly in maintaining student emotional engagement and addressing diverse, dynamic learning needs. This research proposes a hybrid adaptive learning framework to improve the students’ emotional engagement and optimize knowledge acquisition within an Intelligent Tutoring System (ITS). The proposed framework leverages Knowledge Space Theory (KST) to provide a structured theoretical foundation for modelling students’ knowledge as a hierarchical space of concepts and their prerequisites. Partially Observable Markov Decision Processes (POMDP) facilitate adaptive decision-making in ITS, even under uncertainty, guiding the students through logical skills acquisition aligned with KST pathways. Furthermore, Q-learning is employed to implement a Dynamic Learning Pathway (DLP) aimed at improving personalized learning experiences. The proposed approach supports mastery of concepts and fosters emotional engagement by maintaining an optimal balance between difficulty and progression, helping students stay motivated and emotionally invested in their learning journey.

 

Keywords:

Intelligent Tutoring Systems, Knowledge Space Theory, POMDP, Q-learning, Emotional Engagement

 

Thesis Committee:

Internal Reader: Olena Syrotkina, Pooya Moradian Zadeh             

External Reader: Leo Oriet          

Advisor: Xiaobu Yuan

Chair: Jessica Chen

Vector Logo

 

Registration Link (Only MAC students need to pre-register)