A Review of Sentiment Analysis Approaches for Quality Assurance in Teaching and Learning (RETRACTED)

Abstract

The education industry considers quality to be a crucial factor in its development. Nevertheless, the quality of many institutions is far from perfect, as there is a high rate of systemic failure and low performance among students. Consequently, the application of digital computing plays an increasingly important role in assuring the overall quality of an educational institution. However, the literature lacks a reasonable number of systematic reviews that classify research that applied natural language processing and machine learning solutions for students’ sentiment analysis and quality assurance feedback. Thus, this paper presents a systematic literature review that structure available published papers between 2014 and 2023 in a high-impact journal-indexed database. The work extracted 59 relevant papers from the 3392 initially found using exclusion and inclusion criteria. The result identified five (5) prevalent techniques that are majorly researched for sentiment analysis in education and the prevalent supervised machine learning algorithms, lexicon-based approaches, and evaluation metrics in assessing feedback in the education domain.

Empowering educational policy management: understanding student emotions and sentiments

“…However, the SA and emotion analysis results help teachers and universities take corrective measures. Oghu et al (2022) examined 59 papers from an initial pool of 3,392 on NLP techniques and education quality from 2004 to 2023. While lexicon-based learning approaches are frequently associated with supervised or unsupervised learning techniques, more research needs to be done on using students' emotional expressions to provide feedback for student SA.…”

“…Oghu et al. (2022) examined 59 papers from an initial pool of 3,392 on NLP techniques and education quality from 2004 to 2023.…”