Abstract:The rapid development of Massive Open Online Courses (MOOCs) platforms has created an urgent need for an efficient personalized course recommender system that can assist learners of all backgrounds and levels of knowledge in selecting appropriate courses. Currently, most existing methods utilize a sequential recommendation paradigm that captures the user’s learning interests from their learning history, typically through recurrent or graph neural networks. However, fewer studies have explored how to incorporate principles of human learning at both the course and category levels to enhance course recommendations. In this article, we aim at addressing this gap by introducing a novel model, named Prerequisite-Enhanced Catory-Aware Graph Neural Network (PCGNN), for course recommendation. Specifically, we first construct a course prerequisite graph that reflects the human learning principles and further pre-train the course prerequisite relationships as the base embeddings for courses and categories. Then, to capture the user’s complex learning patterns, we build an item graph and a category graph from the user’s historical learning records, respectively: (1) the item graph reflects the course-level local learning transition patterns and (2) the category graph provides insight into the user’s long-term learning interest. Correspondingly, we propose a user interest encoder that employs a gated graph neural network to learn the course-level user interest embedding and design a category transition pattern encoder that utilizes GRU to yield the category-level user interest embedding. Finally, the two fine-grained user interest embeddings are fused to achieve precise course prediction. Extensive experiments on two real-world datasets demonstrate the effectiveness of PCGNN compared with other state-of-the-art methods.

Inferring Concept Prerequisite Relations from Online Educational Resources

Prerequisite Relation Learning for Concepts in Moocs

Concept Extraction and Prerequisite Relation Learning from Educational Data.

A Learning-path Based Supervised Method for Concept Prerequisite Relations Extraction in Educational Data

Heterogeneous Graph Neural Networks for Concept Prerequisite Relation Learning in Educational Data.

Pool-based Active Learning Framework for Concept Prerequisite Learning

Measuring Prerequisite Relations among Concepts.

Exploring the Effectiveness of Student Behavior in Prerequisite Relation Discovery for Concepts

R-VGAE: Relational-variational Graph Autoencoder for Unsupervised Prerequisite Chain Learning

A BERT-Based Approach for Extracting Prerequisite Relations among Wikipedia Concepts

What Should I Learn First: Introducing LectureBank for NLP Education and Prerequisite Chain Learning

Active Learning for Concept Prerequisite Learning in Wikipedia.

A Prerequisite Attention Model for Knowledge Proficiency Diagnosis of Students

Prerequisite-Enhanced Category-Aware Graph Neural Networks for Course Recommendation

Learning Permutation-Invariant Embeddings for Description Logic Concepts

Concept Prerequisite Relation Prediction by Using Permutation-Equivariant Directed Graph Neural Networks

Collaborative Structure Learning Framework from Wikipedia Based on Heterogeneous Graph Attention Network

The curriculum prerequisite network: a tool for visualizing and analyzing academic curricula

Latent Skill Embedding for Personalized Lesson Sequence Recommendation

Prerequisite knowledge‐based automated course planning with semantics consideration

Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation