Search: id:"swepub:oai:DiVA.org:lnu-106625" >
The Potential of Ma...
The Potential of Machine Learning Algorithms for Sentiment Classification of Students’ Feedback on MOOC
-
- Edalati, Maryam (author)
- Norwegian University of Science and Technology, Norway
-
- Imran, Ali Shariq (author)
- Norwegian University of Science and Technology, Norway
-
- Kastrati, Zenun, 1984- (author)
- Linnéuniversitetet,Institutionen för informatik (IK),Institutionen för datavetenskap och medieteknik (DM)
-
show more...
-
- Daudpota, Sher Muhammad (author)
- Sukkur IBA University, Pakistan
-
show less...
-
(creator_code:org_t)
- 2021-08-07
- 2022
- English.
-
In: Intelligent Systems and Applications. - Cham : Springer. - 9783030821982 - 9783030821999 ; , s. 11-22
- Related links:
-
https://urn.kb.se/re...
-
show more...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- Students’ feedback assessment became a hot topic in recent years with growing e-learning platforms coupled with an ongoing pandemic outbreak. Many higher education institutes were compelled to shift on-campus physical classes to online mode, utilizing various online teaching tools and massive open online courses (MOOCs). For many institutes, including both teachers and students, it was a unique and challenging experience conducting lectures and taking classes online. Therefore, analyzing students’ feedback in this crucial time is inevitable for effective teaching and monitoring learning outcomes. Thus, in this paper, we propose and conduct a study to evaluate various machine learning models for aspect-based opinion mining to address this challenge effectively. The proposed approach is trained and validated on a large-scale dataset consisting of manually labeled students’ comments collected from the Coursera online platform. Various conventional machine learning algorithms, namely Random Forest (RF), Support Vector Machine (SVM), and Decision Tree (DT), along with deep-learning methods, are employed to identify teaching-related aspects and predict opinions/attitudes of students towards those aspects. The obtained results are very promising, with an F1 score of 98.01% and 99.43% achieved from RF on the aspect identification and the aspect sentiment classification task, respectively.
Subject headings
- NATURVETENSKAP -- Data- och informationsvetenskap -- Språkteknologi (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Language Technology (hsv//eng)
Keyword
- Informatik
- Information Systems
Publication and Content Type
- ref (subject category)
- kon (subject category)
Find in a library
To the university's database