Cold Start Problem: An Experimental Study of Knowledge Tracing Models with New Students

TL;DR

This study evaluates how different knowledge tracing models perform when predicting the knowledge states of new students with minimal prior data, highlighting current limitations and the need for more generalizable models.

Contribution

It introduces an experimental framework focusing on cold start scenarios, assessing multiple KT models trained on historical data and tested on entirely new students.

Findings

All models struggle initially under cold start conditions.

Performance improves as more student interactions are observed.

SAKT achieves higher initial accuracy but still has limitations.

Abstract

KnowledgeTracing (KT) involves predicting students' knowledge states based on their interactions with Intelligent Tutoring Systems (ITS). A key challenge is the cold start problem, accurately predicting knowledge for new students with minimal interaction data. Unlike prior work, which typically trains KT models on initial interactions of all students and tests on their subsequent interactions, our approach trains models solely using historical data from past students, evaluating their performance exclusively on entirely new students. We investigate cold start effects across three KT models: Deep Knowledge Tracing (DKT), Dynamic Key-Value Memory Networks (DKVMN), and Self-Attentive Knowledge Tracing (SAKT), using ASSISTments 2009, 2015, and 2017 datasets. Results indicate all models initially struggle under cold start conditions but progressively improve with more interactions; SAKT shows higher initial accuracy yet still faces limitations. These findings highlight the need for KT models that effectively generalize to new learners, emphasizing the importance of developing models robust in few-shot and zero-shot learning scenarios