Studying the relationship between BKT fitting error and the skill difficulty index

Francesc Martori, Jordi Cuadros, González Sabaté Lucinio

Producció científica: Capítol de llibreContribució a congrés/conferènciaAvaluat per experts

1 Citació (Scopus)

Resum

Bayesian Knowledge Tracing (BKT) is one of the most popular knowledge inference models due to its interpretability and ability to infer student knowledge. A proper student modeling can help guide the behavior of a cognitive tutor system and provide insight to researchers on understanding how students learn. Using four different datasets we study the relationship between the error coming from fitting the parameters and the difficulty index of the skills and the effect of the size of the dataset in this relationship. The relationship between the fitting error and the difficulty index can be very easy modeled and might be indicating some problems with BKTs performance. However, large datasets are required to clearly see this connection as there is an important sample size effect.

Idioma originalAnglès
Títol de la publicacióLAK 2016 Conference Proceedings, 6th International Learning Analytics and Knowledge Conference - Enhancing Impact
Subtítol de la publicacióConvergence of Communities for Grounding, Implementation, and Validation
EditorAssociation for Computing Machinery
Pàgines364-368
Nombre de pàgines5
ISBN (electrònic)9781450341905
DOIs
Estat de la publicacióPublicada - 25 d’abr. 2016
Esdeveniment6th International Conference on Learning Analytics and Knowledge, LAK 2016 - Edinburgh, United Kingdom
Durada: 25 d’abr. 201629 d’abr. 2016

Sèrie de publicacions

NomACM International Conference Proceeding Series
Volum25-29-April-2016

Conferència

Conferència6th International Conference on Learning Analytics and Knowledge, LAK 2016
País/TerritoriUnited Kingdom
CiutatEdinburgh
Període25/04/1629/04/16

Fingerprint

Navegar pels temes de recerca de 'Studying the relationship between BKT fitting error and the skill difficulty index'. Junts formen un fingerprint únic.

Com citar-ho