Skip to content

Semi-supervised deep learning bias-adjustment framework for IRT parameter estimation under nonignorable missingness

Research Method · All research methods
Evidence · 1 claim (1 for) · 1 study (1 design), q1 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The article introduces a framework combining semi-supervised learning with deep learning to adjust biased 2PL-IRT parameter estimates caused by nonignorable missingness in VLE data. A deep feedforward network with 3 hidden layers and ReLU activation converts observed response patterns of anchor students into unbiased ability estimates, trained by minimizing a weighted cost function combining MSE against biased 2PL-IRT ability estimates and cross-entropy against observed response patterns. The authors state that "the idea of semi-supervised learning was first time used in IRT area to improve the robustness of latent trait estimation", with hyperparameters determined by the elbow method in vali

Accounts

  • Semi-supervised deep learning bias-adjustment framework for IRT parameter estimation under nonignorable missingness: The article introduces a framework combining semi-supervised learning with deep learning to adjust biased 2PL-IRT parameter estimates caused by nonignorable missingness in VLE data. A deep feedforward network with 3 hidden layers and ReLU activation converts observed response patterns of anchor students into unbiased ability estimates, trained by minimizing a weighted cost function combining MSE against biased 2PL-IRT ability estimates and cross-entropy against observed response patterns. The authors state that "the idea of semi-supervised learning was first time used in IRT area to improve the robustness of latent trait estimation", with hyperparameters determined by the elbow method in vali (Xue et al. (2020))

Claims

-

Key Sources

  • Xue, K., Huggins-Manley, A. C., & Leite, W. (2020). Semi-supervised Learning Method for Adjusting Biased Item Difficulty Estimates Caused by Nonignorable Missingness under 2PL-IRT Model. Proceedings of The 13th International Conference on Educational Data Mining (EDM 2020), pp. 715-719. https://educationaldatamining.org/EDM2020/
Page metadata
Field Value
type research-method
id semi-supervised-deep-learning-bias-adjustment-framework-for-irt-parameter-estimation-under
title Semi-supervised deep learning bias-adjustment framework for IRT parameter estimation under nonignorable missingness
description The article introduces a framework combining semi-supervised learning with deep learning to adjust biased 2PL-IRT parameter estimates caused by nonignorable missingness in VLE data.
status review
moved_from theory/semi-supervised-deep-learning-irt-bias-adjustment-framework
generated by process:sweep-kinds on 2026-10-10
trust unverified — no one has confirmed this page's content
sources 1 — xue-2020

Frontmatter is the page's machine-readable record — see the Schema & Guide.