Enhancing Learning Path Recommendation Via Multi-Task Learning

Author

• Afsana Nasrin, Prairie View A&M University

Date of Award

12-2025

Document Type

Thesis

Degree Name

Master of Science

Department

Electrical Engineering

Abstract

Personalized learning is a student-centered educational approach that acknowledges and adapts to the unique characteristics of each learner by customizing content, pacing, and assessments to meet individual needs. Key components include learner profiles, cognitive styles, learning objectives, adjustable pacing, and enabling technologies such as intelligent learning environments, intelligent tutoring systems, and data mining.

A key strategy for implementing personalized learning is learning path recommendation, a technique that guides learners through tailored sequences of content. This approach sequentially recommends personalized learning items, such as lectures, exercises, to address each student’s specific needs. Recent advances in deep learning, particularly deep reinforcement learning, have made learning path recommendation more practical and expressive. However, most of the existing methods generally focus solely on learning path recommendation without fully considering the contributions of related tasks, such as knowledge tracing.

This thesis proposed a multi-task learning model to enhance learning path recommendation by leveraging shared information across multiple related tasks. The approach reframes learning path recommendation as a sequence-to-sequence (Seq2Seq) prediction problem, generating a personalized learning path based on a learner’s historical interactions. The model employed multi-task deep learning to jointly optimize two tasks: learning path recommendation and deep knowledge tracing. Its architecture consisted of a shared Long Short-Term Memory (LSTM) layer to capture common features and two task-specific LSTM layers to model the unique objectives of each task. To mitigate redundant recommendations, a non-repeated loss was incorporated to penalize repeated items in the generated learning path. Extensive experiments on the ASSIST09 dataset demonstrated that the proposed model substantially improved the effectiveness of learning path recommendation.

Index Terms — Deep knowledge tracing (DKT), learning path recommendation (LPR), Multi-task learning (MTL), Personalized learning (PL).

Committee Chair/Advisor

Xishuang Dong

Committee Member

Lin Gong

Committee Member

Anne Lippert

Publisher

Prairie View A&M University

Rights

Date of Digitization

6/22/2026

Contributing Institution

John B Coleman Library

City of Publication

Prairie View

MIME Type

Application/PDF

Recommended Citation

Nasrin, A. (2025). Enhancing Learning Path Recommendation Via Multi-Task Learning. Retrieved from https://digitalcommons.pvamu.edu/pvamu-theses/1677