NEHYBA, Jan and Michal ŠTEFÁNIK. Applications of deep language models for reflective writings. Education and Information Technologies. UNITED STATES: SPRINGER, 2023, vol. 28, No 3, p. 2961-2999. ISSN 1360-2357. Available from: https://dx.doi.org/10.1007/s10639-022-11254-7.
Other formats:   BibTeX LaTeX RIS
Basic information
Original name Applications of deep language models for reflective writings
Authors NEHYBA, Jan (203 Czech Republic, guarantor, belonging to the institution) and Michal ŠTEFÁNIK (703 Slovakia, belonging to the institution).
Edition Education and Information Technologies, UNITED STATES, SPRINGER, 2023, 1360-2357.
Other information
Original language English
Type of outcome Article in a journal
Field of Study 50301 Education, general; including training, pedagogy, didactics [and education systems]
Country of publisher United States of America
Confidentiality degree is not subject to a state or trade secret
WWW URL
Impact factor Impact factor: 5.500 in 2022
RIV identification code RIV/00216224:14330/23:00129992
Organization unit Faculty of Informatics
Doi http://dx.doi.org/10.1007/s10639-022-11254-7
UT WoS 000849683800001
Keywords in English Deep learning; Natural language processing; Reflection dataset; Reflection classification; Analyses of reflective journals; Generalized linear mixed models
Tags International impact, Reviewed
Changed by Changed by: Mgr. Daniela Marcollová, učo 111148. Changed: 22/2/2024 15:58.
Abstract
Social sciences expose many cognitively complex, highly qualified, or fuzzy problems, whose resolution relies primarily on expert judgement rather than automated systems. One of such instances that we study in this work is a reflection analysis in the writings of student teachers. We share a hands-on experience on how these challenges can be successfully tackled in data collection for machine learning. Based on the novel deep learning architectures pre-trained for a general language understanding, we can reach an accuracy ranging from 76.56–79.37% on low-confidence samples to 97.56–100% on high confidence cases. We open-source all our resources and models, and use the models to analyse previously ungrounded hypotheses on reflection of university students. Our work provides a toolset for objective measurements of reflection in higher education writings, applicable in more than 100 other languages worldwide with a loss in accuracy measured between 0–4.2% Thanks to the outstanding accuracy of the deep models, the presented toolset allows for previously unavailable applications, such as providing semi-automated student feedback or measuring an effect of systematic changes in the educational process via the students’ response.
Links
MUNI/A/1339/2022, interní kód MUName: Rozvoj technik pro zpracování dat pro podporu vyhledávání, analýz a vizualizací rozsáhlých datových souborů s využitím umělé inteligence
Investor: Masaryk University, Development of data processing techniques to support search, analysis and visualization of large datasets using artificial intelligence
PrintDisplayed: 23/7/2024 02:37