DiAD: Domain adaptation for learning at scale

Ziheng Zeng, Suma Pallathadka Bhat, Snigdha Chaturvedi, Dan Roth

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Massive online courses occupy an important place in the educational landscape of today. We study an approach to scale predictive analytic models derived from online course discussion fora-specifically that of confusion detection-onto other courses. The primary challenge here is the lack of labeled examples in a new course and this calls for unsupervised domain adaptation (DA). As a first step in exploring DA in the education domain, we propose a simple algorithm, DiAd, which adapts a classifier trained on a course with labeled data by selectively choosing instances from a new course (with no labeled data) that are most dissimilar to the course with labeled data and on which the classifier is very confident of classification. Our algorithm is empirically validated on the confusion detection task across multiple online courses. We find that DiAd outperforms other methods on the target domain, while showing a comparable performance to a popular method that uses labeled data from the target domain.

Original languageEnglish (US)
Title of host publicationProceedings of the 9th International Conference on Learning Analytics and Knowledge
Subtitle of host publicationLearning Analytics to Promote Inclusion and Success, LAK 2019
PublisherAssociation for Computing Machinery
Pages185-194
Number of pages10
ISBN (Electronic)9781450362566
DOIs
StatePublished - Mar 4 2019
Event9th International Conference on Learning Analytics and Knowledge, LAK 2019 - Tempe, United States
Duration: Mar 4 2019Mar 8 2019

Publication series

NameACM International Conference Proceeding Series

Conference

Conference9th International Conference on Learning Analytics and Knowledge, LAK 2019
CountryUnited States
CityTempe
Period3/4/193/8/19

Fingerprint

Classifiers
Education
Predictive analytics

Keywords

  • Confusion detection
  • Domain adaptation
  • Learning at scale

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Cite this

Zeng, Z., Bhat, S. P., Chaturvedi, S., & Roth, D. (2019). DiAD: Domain adaptation for learning at scale. In Proceedings of the 9th International Conference on Learning Analytics and Knowledge: Learning Analytics to Promote Inclusion and Success, LAK 2019 (pp. 185-194). (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3303772.3303810

DiAD : Domain adaptation for learning at scale. / Zeng, Ziheng; Bhat, Suma Pallathadka; Chaturvedi, Snigdha; Roth, Dan.

Proceedings of the 9th International Conference on Learning Analytics and Knowledge: Learning Analytics to Promote Inclusion and Success, LAK 2019. Association for Computing Machinery, 2019. p. 185-194 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zeng, Z, Bhat, SP, Chaturvedi, S & Roth, D 2019, DiAD: Domain adaptation for learning at scale. in Proceedings of the 9th International Conference on Learning Analytics and Knowledge: Learning Analytics to Promote Inclusion and Success, LAK 2019. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 185-194, 9th International Conference on Learning Analytics and Knowledge, LAK 2019, Tempe, United States, 3/4/19. https://doi.org/10.1145/3303772.3303810
Zeng Z, Bhat SP, Chaturvedi S, Roth D. DiAD: Domain adaptation for learning at scale. In Proceedings of the 9th International Conference on Learning Analytics and Knowledge: Learning Analytics to Promote Inclusion and Success, LAK 2019. Association for Computing Machinery. 2019. p. 185-194. (ACM International Conference Proceeding Series). https://doi.org/10.1145/3303772.3303810
Zeng, Ziheng ; Bhat, Suma Pallathadka ; Chaturvedi, Snigdha ; Roth, Dan. / DiAD : Domain adaptation for learning at scale. Proceedings of the 9th International Conference on Learning Analytics and Knowledge: Learning Analytics to Promote Inclusion and Success, LAK 2019. Association for Computing Machinery, 2019. pp. 185-194 (ACM International Conference Proceeding Series).
@inproceedings{e336ea77902c4c69bc285bd0ccdf4133,
title = "DiAD: Domain adaptation for learning at scale",
abstract = "Massive online courses occupy an important place in the educational landscape of today. We study an approach to scale predictive analytic models derived from online course discussion fora-specifically that of confusion detection-onto other courses. The primary challenge here is the lack of labeled examples in a new course and this calls for unsupervised domain adaptation (DA). As a first step in exploring DA in the education domain, we propose a simple algorithm, DiAd, which adapts a classifier trained on a course with labeled data by selectively choosing instances from a new course (with no labeled data) that are most dissimilar to the course with labeled data and on which the classifier is very confident of classification. Our algorithm is empirically validated on the confusion detection task across multiple online courses. We find that DiAd outperforms other methods on the target domain, while showing a comparable performance to a popular method that uses labeled data from the target domain.",
keywords = "Confusion detection, Domain adaptation, Learning at scale",
author = "Ziheng Zeng and Bhat, {Suma Pallathadka} and Snigdha Chaturvedi and Dan Roth",
year = "2019",
month = "3",
day = "4",
doi = "10.1145/3303772.3303810",
language = "English (US)",
series = "ACM International Conference Proceeding Series",
publisher = "Association for Computing Machinery",
pages = "185--194",
booktitle = "Proceedings of the 9th International Conference on Learning Analytics and Knowledge",

}

TY - GEN

T1 - DiAD

T2 - Domain adaptation for learning at scale

AU - Zeng, Ziheng

AU - Bhat, Suma Pallathadka

AU - Chaturvedi, Snigdha

AU - Roth, Dan

PY - 2019/3/4

Y1 - 2019/3/4

N2 - Massive online courses occupy an important place in the educational landscape of today. We study an approach to scale predictive analytic models derived from online course discussion fora-specifically that of confusion detection-onto other courses. The primary challenge here is the lack of labeled examples in a new course and this calls for unsupervised domain adaptation (DA). As a first step in exploring DA in the education domain, we propose a simple algorithm, DiAd, which adapts a classifier trained on a course with labeled data by selectively choosing instances from a new course (with no labeled data) that are most dissimilar to the course with labeled data and on which the classifier is very confident of classification. Our algorithm is empirically validated on the confusion detection task across multiple online courses. We find that DiAd outperforms other methods on the target domain, while showing a comparable performance to a popular method that uses labeled data from the target domain.

AB - Massive online courses occupy an important place in the educational landscape of today. We study an approach to scale predictive analytic models derived from online course discussion fora-specifically that of confusion detection-onto other courses. The primary challenge here is the lack of labeled examples in a new course and this calls for unsupervised domain adaptation (DA). As a first step in exploring DA in the education domain, we propose a simple algorithm, DiAd, which adapts a classifier trained on a course with labeled data by selectively choosing instances from a new course (with no labeled data) that are most dissimilar to the course with labeled data and on which the classifier is very confident of classification. Our algorithm is empirically validated on the confusion detection task across multiple online courses. We find that DiAd outperforms other methods on the target domain, while showing a comparable performance to a popular method that uses labeled data from the target domain.

KW - Confusion detection

KW - Domain adaptation

KW - Learning at scale

UR - http://www.scopus.com/inward/record.url?scp=85062779859&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062779859&partnerID=8YFLogxK

U2 - 10.1145/3303772.3303810

DO - 10.1145/3303772.3303810

M3 - Conference contribution

AN - SCOPUS:85062779859

T3 - ACM International Conference Proceeding Series

SP - 185

EP - 194

BT - Proceedings of the 9th International Conference on Learning Analytics and Knowledge

PB - Association for Computing Machinery

ER -