Enhancing Student Performance Prediction on Learnersourced Questions with SGNN-LLM Synergy

TL;DR

This paper introduces a novel approach combining Signed Graph Neural Networks and Large Language Model embeddings to improve student performance prediction on learnersourced questions, especially under cold start conditions, validated across multiple real-world datasets.

Contribution

The paper presents an innovative synergy of SGNNs and LLMs for better performance prediction in noisy and sparse data scenarios, addressing cold start challenges in learnersourced education.

Findings

Outperforms baseline methods in predictive accuracy.

Enhances robustness against noisy and sparse data.

Effective across five real-world datasets.

Abstract

Learnersourcing offers great potential for scalable education through student content creation. However, predicting student performance on learnersourced questions, which is essential for personalizing the learning experience, is challenging due to the inherent noise in student-generated data. Moreover, while conventional graph-based methods can capture the complex network of student and question interactions, they often fall short under cold start conditions where limited student engagement with questions yields sparse data. To address both challenges, we introduce an innovative strategy that synergizes the potential of integrating Signed Graph Neural Networks (SGNNs) and Large Language Model (LLM) embeddings. Our methodology employs a signed bipartite graph to comprehensively model student answers, complemented by a contrastive learning framework that enhances noise resilience. Furthermore, LLM's contribution lies in generating foundational question embeddings, proving especially advantageous in addressing cold start scenarios characterized by limited graph data. Validation across five real-world datasets sourced from the PeerWise platform underscores our approach's effectiveness. Our method outperforms baselines, showcasing enhanced predictive accuracy and robustness.