Skip to content

Logistic Knowledge Tracing Models (LFA, PFA, KTM)

Theory · All theories
Evidence · no claims cited

Description

The survey's logistic branch covers Learning Factor Analysis, Performance Factor Analysis and Knowledge Tracing Machines. "Logistic models represent the probability of students cor- rectly answering exercises as a logistic function of the student and KC parameters." LFA uses initial knowledge, KC easiness and KC learning rate; PFA uses prior failures, prior successes and KC easiness; KTM uses factorization machines to encode side information about exercises or students.

Design Implications

Context

Requirements

  • Factors estimated from students' learning interactions (e.g., prior successes and failures on a KC, KC easiness) combined through a logistic (sigmoid) function

Constraints

  • The survey notes the original PFA model ignores the order of answers and the time between learning interactions, making it difficult to incorporate time information directly

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

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 Logistic Knowledge Tracing Models (LFA, PFA, KTM)
description The survey's logistic branch covers Learning Factor Analysis, Performance Factor Analysis and Knowledge Tracing Machines.
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.