The impact of automatic text translation on classification of online discussions for social and cognitive presences

Arthur Barbosa, Máverick Ferreira, Rafael Ferreira Mello, Rafael Dueire Lins, Dragan Gasevic

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

1 Citation (Scopus)

Abstract

This paper reports the findings of a study that measured the effectiveness of employing automatic text translation methods in automated classification of online discussion messages according to the categories of social and cognitive presences. Specifically, we examined the classification of 1,500 Portuguese and 1,747 English discussion messages using classifiers trained on the datasets before and after the application of text translation. While the English model generated, with the original and translated texts, achieved results (accuracy and Cohen's κ) similar to those of the previously reported studies, the translation to Portuguese led to a decrease in the performance. The indicates the general viability of the proposed approach when converting the text to English. Moreover, this study highlighted the importance of different features and resources, and the limitations of the resources for Portuguese as reasons of the results obtained.

Original languageEnglish
Title of host publicationLAK21 Conference Proceedings - The Impact we Make: The contributions of learning analytics to learning
Subtitle of host publicationThe Eleventh International Conference on Learning Analytics & Knowledge
EditorsMaren Scheffel, Nia Dowell, Srecko Joksimovic, George Siemens
Place of PublicationNew York NY USA
PublisherAssociation for Computing Machinery (ACM)
Pages77-87
Number of pages11
ISBN (Electronic)9781450389358
DOIs
Publication statusPublished - 2021
EventInternational Learning Analytics & Knowledge Conference 2021 - Online, Irvine, United States of America
Duration: 12 Apr 202116 Apr 2021
Conference number: 11th
https://www.solaresearch.org/events/lak/lak21/
https://dl.acm.org/doi/proceedings/10.1145/3448139 (Proceedings)

Conference

ConferenceInternational Learning Analytics & Knowledge Conference 2021
Abbreviated titleLAK 2021
CountryUnited States of America
CityIrvine
Period12/04/2116/04/21
Internet address

Keywords

  • Community of inquiry model
  • Content analytics
  • Online discussion
  • Text translation

Cite this