Detection of Student Modelling Anomalies

Sergey Sosnovsky*, Laurens Müter, Marc Valkenier, Matthieu Brinkhuis, Abe Hofman

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

As the modern TEL tools gain wider adoption in real educational contexts, they start facing important practical problems. One such problem for adaptive educational systems is the reliability of their student modelling mechanisms. Even when such a mechanism has been tested and calibrated to represent students’ knowledge reasonably well, the student herself can become a source of problems. Students can use the system in a non-intended way, exhibit long periods of off task behaviour, try gaming the system, seek help of parents or peers, etc. Such usage patterns will manifest themselves in sequences of activity that do not represent student abilities and will result in student modelling anomalies causing subsequent suboptimal adaptive interventions from the system. This would be very important for a system that is used in real classrooms with younger children, especially, when it is also available at home as a supporting tool for independent work. This paper reports a study of such a system – Math Garden. Several user modelling anomalies have been detected in its logs. First steps towards building an automated tool for on-the-fly student modelling anomaly detection are reported.

Original languageEnglish
Title of host publicationLifelong Technology-Enhanced Learning - 13th European Conference on Technology Enhanced Learning, EC-TEL 2018, Proceedings
PublisherSpringer
Pages531-536
Number of pages6
ISBN (Print)9783319985718
DOIs
Publication statusPublished - 1 Jan 2018
Event13th European Conference on Technology Enhanced Learning, EC-TEL 2018 - Leeds, United Kingdom
Duration: 3 Sept 20186 Sept 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11082 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference13th European Conference on Technology Enhanced Learning, EC-TEL 2018
Country/TerritoryUnited Kingdom
CityLeeds
Period3/09/186/09/18

Keywords

  • Adaptive educational system
  • Educational data mining
  • Student modelling
  • Student modelling anomaly

Fingerprint

Dive into the research topics of 'Detection of Student Modelling Anomalies'. Together they form a unique fingerprint.

Cite this