Predicting the Difficulty of Exercise Items for Dynamic Difficulty Adaptation in Adaptive Language Tutoring

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

Predicting the Difficulty of Exercise Items for Dynamic Difficulty Adaptation in Adaptive Language Tutoring. / Pandarova, Irina; Schmidt, Torben; Hartig, Johannes et al.
In: International Journal of Artificial Intelligence in Education, Vol. 29, No. 3, 15.08.2019, p. 342-367.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Bibtex

@article{348f671ee47740708ec089a02c9c14c2,
title = "Predicting the Difficulty of Exercise Items for Dynamic Difficulty Adaptation in Adaptive Language Tutoring",
abstract = "Advances in computer technology and artificial intelligence create opportunities for developing adaptive language learning technologies which are sensitive to individual learner characteristics. This paper focuses on one form of adaptivity in which the difficulty of learning content is dynamically adjusted to the learner{\textquoteright}s evolving language ability. A pilot study is presented which aims to advance the (semi-)automatic difficulty scoring of grammar exercise items to be used in dynamic difficulty adaptation in an intelligent language tutoring system for practicing English tenses. In it, methods from item response theory and machine learning are combined with linguistic item analysis in order to calibrate the difficulty of an initial exercise pool of cued gap-filling items (CGFIs) and isolate CGFI features predictive of item difficulty. Multiple item features at the gap, context and CGFI levels are tested and relevant predictors are identified at all three levels. Our pilot regression models reach encouraging prediction accuracy levels which could, pending additional validation, enable the dynamic selection of newly generated items ranging from moderately easy to moderately difficult. The paper highlights further applications of the proposed methodology in the area of adapting language tutoring, item design and second language acquisition, and sketches out issues for future research.",
keywords = "Didactics of English as a foreign language, adaptivity, Intelligent language tutoring systems, Idem difficulty prediction, Item response theory, Machine Learning, Scon language aquisition",
author = "Irina Pandarova and Torben Schmidt and Johannes Hartig and Ahc{\`e}ne Boubekki and Jones, {Roger Dale} and Ulf Brefeld",
note = "doi.org/10.1007/s40593-019-00180-4",
year = "2019",
month = aug,
day = "15",
doi = "10.1007/s40593-019-00180-4",
language = "English",
volume = "29",
pages = "342--367",
journal = "International Journal of Artificial Intelligence in Education",
issn = "1560-4306",
publisher = "Springer Netherlands",
number = "3",

}

RIS

TY - JOUR

T1 - Predicting the Difficulty of Exercise Items for Dynamic Difficulty Adaptation in Adaptive Language Tutoring

AU - Pandarova, Irina

AU - Schmidt, Torben

AU - Hartig, Johannes

AU - Boubekki, Ahcène

AU - Jones, Roger Dale

AU - Brefeld, Ulf

N1 - doi.org/10.1007/s40593-019-00180-4

PY - 2019/8/15

Y1 - 2019/8/15

N2 - Advances in computer technology and artificial intelligence create opportunities for developing adaptive language learning technologies which are sensitive to individual learner characteristics. This paper focuses on one form of adaptivity in which the difficulty of learning content is dynamically adjusted to the learner’s evolving language ability. A pilot study is presented which aims to advance the (semi-)automatic difficulty scoring of grammar exercise items to be used in dynamic difficulty adaptation in an intelligent language tutoring system for practicing English tenses. In it, methods from item response theory and machine learning are combined with linguistic item analysis in order to calibrate the difficulty of an initial exercise pool of cued gap-filling items (CGFIs) and isolate CGFI features predictive of item difficulty. Multiple item features at the gap, context and CGFI levels are tested and relevant predictors are identified at all three levels. Our pilot regression models reach encouraging prediction accuracy levels which could, pending additional validation, enable the dynamic selection of newly generated items ranging from moderately easy to moderately difficult. The paper highlights further applications of the proposed methodology in the area of adapting language tutoring, item design and second language acquisition, and sketches out issues for future research.

AB - Advances in computer technology and artificial intelligence create opportunities for developing adaptive language learning technologies which are sensitive to individual learner characteristics. This paper focuses on one form of adaptivity in which the difficulty of learning content is dynamically adjusted to the learner’s evolving language ability. A pilot study is presented which aims to advance the (semi-)automatic difficulty scoring of grammar exercise items to be used in dynamic difficulty adaptation in an intelligent language tutoring system for practicing English tenses. In it, methods from item response theory and machine learning are combined with linguistic item analysis in order to calibrate the difficulty of an initial exercise pool of cued gap-filling items (CGFIs) and isolate CGFI features predictive of item difficulty. Multiple item features at the gap, context and CGFI levels are tested and relevant predictors are identified at all three levels. Our pilot regression models reach encouraging prediction accuracy levels which could, pending additional validation, enable the dynamic selection of newly generated items ranging from moderately easy to moderately difficult. The paper highlights further applications of the proposed methodology in the area of adapting language tutoring, item design and second language acquisition, and sketches out issues for future research.

KW - Didactics of English as a foreign language

KW - adaptivity

KW - Intelligent language tutoring systems

KW - Idem difficulty prediction

KW - Item response theory

KW - Machine Learning

KW - Scon language aquisition

UR - http://www.scopus.com/inward/record.url?scp=85065811581&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/272a803a-842d-38d7-b186-cb3092c08c57/

U2 - 10.1007/s40593-019-00180-4

DO - 10.1007/s40593-019-00180-4

M3 - Journal articles

VL - 29

SP - 342

EP - 367

JO - International Journal of Artificial Intelligence in Education

JF - International Journal of Artificial Intelligence in Education

SN - 1560-4306

IS - 3

ER -

Recently viewed

Publications

  1. Das 'Geheimnis', eine fesselnde Geschichte für Kinder zu schreiben, und die didaktische Bedeutung des erzählten Lebens
  2. Tree and mycorrhizal fungal diversity drive intraspecific and intraindividual trait variation in temperate forests
  3. Saproxylic beetles in the Gartow region of Lower Saxony, a hotspot of invertebrate diversity in north-western Germany
  4. Molecular analysis meets morphology-based systematics-a synthetic approach for Chalarinae (Insecta: Diptera: Pipunculidae)
  5. Bird's Response to Revegetation of Different Structure and Floristics-Are "Restoration Plantings" Restoring Bird Communities?
  6. Mentale Modelle bilden und Textinformationen verstehen. Wie Grundschülerinnen und Grundschüler komplexe Leseaufgaben bewältigen
  7. Two high-mountain burnet moth species (Lepidoptera, Zygaenidae) react differently to the global change drivers climate and land-use
  8. Genetic diversity and population structure of the endangered insect species Carabus variolosus in its western distribution range
  9. Buchbesprechung: Wolfgang Wildfeuer: Kommunikation - Moderation - Mediation. Ein Trainingsprogramm für Schüler und Lehrer. Juventa 2006
  10. Untersuchungen zur sozialen Organisation einer Herde von Liebenthaler Pferden im Biosphärenreservat Flusslandschaft Elbe-Brandenburg
  11. Biogeography meets conservation: the genetic structure of the endangered lycaenid butterfly Lycaena helle (Denis & Schiffermüller, 1775)
  12. Application of titanium dioxide nanoparticles as a photocatalyst for the removal of micropollutants such as pharmaceuticals from water
  13. Eemian landscape response to climatic shifts and evidence for northerly Neanderthal occupation at a palaeolake margin in northern Germany
  14. Phenotypic Plasticity Explains Response Patterns of European Beech (Fagus sylvatica L.) Saplings to Nitrogen Fertilization and Drought Events
  15. Rezension zu: Understanding the city, contemporary and future perspectives, John Eade and Christopher Mele (eds.), Oxford, UK Blackwell, 2002, 384 pp.
  16. Tree species richness strengthens relationships between ants and the functional composition of spider assemblages in a highly diverse forest
  17. Zum Zusammenhang von sportunterrichtsbezogenem Wissen, sportunterrichtlicher Leistung und sprachlichen Fähigkeiten von Schülerinnen und Schülern
  18. At the interface of historical and present-day ecology: ground beetles in woodlands and open habitats in Upper Galilee (Israel) (Coleoptera: Carabidae)
  19. Wie kann die professionelle Reflexion von angehenden Lehrer*innen digital gefördert werden? – Chancen und Grenzen neuer Tools in der Lehrer*innenbildung
  20. (Wie) Nutzen angehende Lehrpersonen ihr schriftsystematisches Wissen in didaktischen Anforderungssituationen des schriftsprachlichen Anfangsunterrichts?
  21. Long-Term Abandonment of Forest Management Has a Strong Impact on Tree Morphology and Wood Volume Allocation Pattern of European Beech (Fagus Sylvatica L.)
  22. Aquatic and terrestrial proxy evidence for Middle Pleistocene palaeolake and lake-shore development at two Lower Palaeolithic sites of Schöningen, Germany
  23. Non-native tree species (Pseudotsuga menziesii) strongly decreases predator biomass and abundance in mixed-species plantations of a tree diversity experiment
  24. Population genetics and ecological niche modelling reveal high fragmentation and potential future extinction of the endangered relict butterfly Lycaena helle