Technical Taxonomy of Fundamental Knowledge Tracing Models

Theory · All theories
Evidence · 2 claims (2 mixed) · 1 study (1 review), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description

The survey's first organizing axis sorts fundamental KT models by technical route: "the proposed taxonomy splits existing KT methods into three categories" — Bayesian models built on probability models, logistic models built on logistic functions, and deep learning models built on neural networks. It further divides the deep learning branch into deep, memory-aware, attentive and graph-based knowledge tracing, according to four types of neural networks.

Design Implications

Context

Requirements

• Bayesian models: Bayesian knowledge tracing and dynamic Bayesian knowledge tracing
• Logistic models: learning factor analysis, performance factor analysis, and knowledge tracing machines
• Deep learning models: deep knowledge tracing (RNN/LSTM), memory-aware knowledge tracing (memory networks), attentive knowledge tracing (self-attention), and graph-based knowledge tracing (graph neural networks)

Constraints

• The survey states that deep learning models remain notoriously difficult to interpret owing to their end-to-end learning strategy

Target Learners

• Students whose exercise-answering interactions are recorded by online learning systems such as ASSISTments, Junyi Academy, Eedi and EdNet's Santa tutor

Target Learning Objectives

• Monitoring students' evolving knowledge states on knowledge concepts and predicting performance on future exercises

Claims

• Deep Learning Knowledge Tracing Outperforms But Lacks Interpretability [~W]
• No Single Knowledge Tracing Model Is Always Best [~W]

Related Theories

• Knowledge Tracing Learner Modeling Task
• Bayesian Knowledge Tracing Two State Model
• Logistic Knowledge Tracing Models
• Adaptive G-UKT: a unified probabilistic knowledge tracing framework coupling Gaussian state representations, adaptive graph topology learning, Wasserstein attention, and uncertainty-gated contrastive regularization

Examples

-

Key Sources

• Shuanghong Shen, Qi Liu, Zhenya Huang, Yonghe Zheng, Minghao Yin, Minjuan Wang, and Enhong Chen. (2021). A Survey of Knowledge Tracing: Models, Variants, and Applications. https://arxiv.org/abs/2105.15106

Page metadata

Field	Value
type	theory
title	Technical Taxonomy of Fundamental Knowledge Tracing Models
description	The survey's first organizing axis sorts fundamental KT models by technical route: "the proposed taxonomy splits existing KT methods into three categories" — Bayesian models built on probability models, logistic model...
status	review
canonical	True
generated	by claude/unspecified on 2026-09-24
trust	unverified — no one has confirmed this page's content
sources	1 — shuanghong-shen-2021

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