Machine Learning for Knowledge Tracing in Learner Models

Colloque du CRIFPE
Communication orale
Le numérique éducatif
Machine Learning (ML) techniques are being intensively applied in educational settings: supervised, unsupervised, semi-supervised and reinforcement learning techniques are used to predict grades and skills, grade exams, recognize meaning in text, evaluate open answers, suggest appropriate educational resources, and group or associate students with similar characteristics or interests. While modeling the learner, Knowledge Tracing (KT) is a significant data feature employed to predict student’s performance, as it tracks the knowledge state of students based on observed outcomes from their previous educational practices, such as answes, grades and/or behaviours. In this study, we review commonly used ML techniques for KT spred in 60+ papers on the topic. We extract discerning characteristics of ML for KT in Learner Models (LM) that will allow us to understand the current panorama of ML applied to KT and help up choose an adequate ML technique for KT in LM. This work is dedicated to MOOC designers/providers, pedagogical engineers and researchers who meet difficulties to apply ML to KT in LM. This project was supported by the French government through the Programme Investissement d’Avenir (I-SITE ULNE / ANR-16-IDEX-0004 ULNE) managed by the Agence Nationale de la Recherche.
  • Sergio Ramirez - Université de Lille - Laboratoire CIREL / Trigone
  • Jean Heutte - Université de Lille - Laboratoire CIREL / Trigone
  • Nour El Mawas - Université de Lille - Laboratoire CIREL / Trigone
Diffusion en direct
Vous devez être connecté pour accéder à la diffusion.
2021-04-30 13 h 00
30 minutes