Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts

Xinjie Sun, Qi Liu, Kai Zhang, Shuanghong Shen, Fei Wang, Yan Zhuang, Zheng Zhang, Weiyin Gong, Shijin Wang, Lina Yang, Xingying Huo

TL;DR

This work addresses the challenge of cognitive diagnosis in education by introducing a Hierarchy Constraint-Aware Cognitive Diagnosis Framework (HCD) that jointly models personalized and hierarchy-driven knowledge states. It features a two-stage architecture with a Hierarchy Mapping layer, intra-level Convolution-Enhanced Attention (CEA), and inter-level Inter-Sampling Attention (RSA), plus a Personalized Diagnostic Enhancement to blend absolute and relative proficiency. The model optimizes a multi-term cross-entropy objective and demonstrates superior predictive performance and interpretability on the PISA 2015 Science, Reading, and Math datasets compared with classical CDMs and their HCD variants. The results suggest hierarchy-aware diagnostics yield more realistic and fair assessments, with implications for scalable, interpretable educational analytics and guidance for future refinement of hierarchy granularity and psychology-informed modeling.

Abstract

Cognitive diagnosis (CD) aims to reveal students' proficiency in specific knowledge concepts. With the increasing adoption of intelligent education applications, accurately assessing students' knowledge mastery has become an urgent challenge. Although existing cognitive diagnosis frameworks enhance diagnostic accuracy by analyzing students' explicit response records, they primarily focus on individual knowledge state, failing to adequately reflect the relative ability performance of students within hierarchies. To address this, we propose the Hierarchy Constraint-Aware Cognitive Diagnosis Framework (HCD), designed to more accurately represent student ability performance within real educational contexts. Specifically, the framework introduces a hierarchy mapping layer to identify students' levels. It then employs a hierarchy convolution-enhanced attention layer for in-depth analysis of knowledge concepts performance among students at the same level, uncovering nuanced differences. A hierarchy inter-sampling attention layer captures performance differences across hierarchies, offering a comprehensive understanding of the relationships among students' knowledge state. Finally, through personalized diagnostic enhancement, the framework integrates hierarchy constraint perception features with existing models, improving the representation of both individual and group characteristics. This approach enables precise inference of students' knowledge state. Research shows that this framework not only reasonably constrains changes in students' knowledge states to align with real educational settings, but also supports the scientific rigor and fairness of educational assessments, thereby advancing the field of cognitive diagnosis.

Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts

TL;DR

This work addresses the challenge of cognitive diagnosis in education by introducing a Hierarchy Constraint-Aware Cognitive Diagnosis Framework (HCD) that jointly models personalized and hierarchy-driven knowledge states. It features a two-stage architecture with a Hierarchy Mapping layer, intra-level Convolution-Enhanced Attention (CEA), and inter-level Inter-Sampling Attention (RSA), plus a Personalized Diagnostic Enhancement to blend absolute and relative proficiency. The model optimizes a multi-term cross-entropy objective and demonstrates superior predictive performance and interpretability on the PISA 2015 Science, Reading, and Math datasets compared with classical CDMs and their HCD variants. The results suggest hierarchy-aware diagnostics yield more realistic and fair assessments, with implications for scalable, interpretable educational analytics and guidance for future refinement of hierarchy granularity and psychology-informed modeling.

Abstract

Cognitive diagnosis (CD) aims to reveal students' proficiency in specific knowledge concepts. With the increasing adoption of intelligent education applications, accurately assessing students' knowledge mastery has become an urgent challenge. Although existing cognitive diagnosis frameworks enhance diagnostic accuracy by analyzing students' explicit response records, they primarily focus on individual knowledge state, failing to adequately reflect the relative ability performance of students within hierarchies. To address this, we propose the Hierarchy Constraint-Aware Cognitive Diagnosis Framework (HCD), designed to more accurately represent student ability performance within real educational contexts. Specifically, the framework introduces a hierarchy mapping layer to identify students' levels. It then employs a hierarchy convolution-enhanced attention layer for in-depth analysis of knowledge concepts performance among students at the same level, uncovering nuanced differences. A hierarchy inter-sampling attention layer captures performance differences across hierarchies, offering a comprehensive understanding of the relationships among students' knowledge state. Finally, through personalized diagnostic enhancement, the framework integrates hierarchy constraint perception features with existing models, improving the representation of both individual and group characteristics. This approach enables precise inference of students' knowledge state. Research shows that this framework not only reasonably constrains changes in students' knowledge states to align with real educational settings, but also supports the scientific rigor and fairness of educational assessments, thereby advancing the field of cognitive diagnosis.

Paper Structure

This paper contains 27 sections, 21 equations, 7 figures, 3 tables.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORKS
  3. Cognitive Diagnosis
  4. Educational Hierarchy
  5. PRELIMINARY
  6. Problem Definition
  7. Basic Factors
  8. HCD Framework
  9. Model Overview
  10. Hierarchy Constraint-Aware Modeling
  11. Hierarchy Mapping
  12. One-hot Embedding
  13. Intra-Hierarchy Convolution Enhanced Attention Layer (CEA)
  14. Inter-Hierarchy Random Sample Attention Layer (RSA)
  15. Personalized Diagnosis Enhancement
  16. ...and 12 more sections

Figures (7)

  • Figure 1: A toy example of cognitive diagnosis, where CDM stands for cognitive diagnosis model and GKP stands for group knowledge proficiency.
  • Figure 2: Overview of HCD Framework.
  • Figure 3: Comparison of kernel density plots for the mean score based on prior statistics, the mean knowledge state of students from the NCDM model, and the mean knowledge state of students from the HCD-NCDM model on the PISA-Read and PISA-Math datasets.
  • Figure 4: Demonstrating knowledge concept Mastery for Students Across Six Levels in the PISA-Math Dataset: Random Selection of One Student from Each Level
  • Figure 5: Comparative analysis of model DOA interpretability across three distinct datasets.
  • ...and 2 more figures