Treffer: Text mining applied to distance higher education: A systematic literature review.

Much of the data produced and consumed by students, teachers, and educational managers is in textual format. Text Mining (TM) and Natural Language Processing (NLP) have been applied in the educational context in different ways. Ideally, such applications combine computational, linguistic, pedagogical, and psychological aspects. This article aims to gather and analyze scientific publications that have applied TM and NLP techniques in textual corpora from distance-higher education through a Systematic Literature Review. Eight scientific databases were searched (ACM DL, Scopus, Web of Science, IEEE Xplore, ArXiv, SpringerLink, ScienceDirect, and ERIC), and publications from 2017 to 2021 were selected. 718 unique publications were screened to identify primary research capable of characterizing this scientific area. 52 resulting publications were fully analyzed, and some consolidated results include: 38% of works had the professors as end users, followed by students (27%) and managers (25%); the English language was present in 50% of publications, followed by the Portuguese language (13,5%) and others languages; the text mining tasks most used were text classification (27%), sentiment analysis (17%), information extraction (15%), chatbot (15%) and topic modeling (13%); LDA (Latent Dirichlet Analysis) was the technique most used (19%); the Python language was the programming language most prevalent (42%), and 54% of works do not mention any educational construct or theory. Thus, this article presents an unprecedented overview of the field of Educational Text Mining (ETM) in distance higher education and analyses the main results obtained, aiming for future research in the area. [ABSTRACT FROM AUTHOR]

Copyright of Education & Information Technologies is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)