Baseline Study of Different Sentiment Analysis Computing Methods to Enhance Quality Assurance In Teaching and Learning

Abstract

The idea of modernization has become a norm in society, and the quality of the education sector that is been delivered to society is worth being assessed, and this will reflect the caliber of graduands these institutions are producing. As the key to this assessment is to x-ray clear-cut metrics used in the assessment of these institutions of higher learning by regulating authorities, parents, and other stakeholders These institutions' academic deliverables in terms of availability of experienced teachers, modern teaching and learning amenities, open access to learning materials, open access to assessment results, and timely responses to transcripts application to mention just a few needs proper evaluation. A lot of institutions' indices of performance are not encouraging, which translates to a high level of students dropping out, poor student performance, and low productivity which affects the standard of living and overall planned change that education is supposed to bring to the generality of the society at large. We must appreciate the efforts of previous researchers who have spent precious time using various sentiment analysis tools to unveil these challenges facing our education institutions and providing far-reaching measures that will make these institutions improve all indexes of measurement approved by regulating authorities globally as a mark of global competitiveness. Literature reviews by researchers using natural language processing and other machine learning techniques to get students and other stakeholders’ feedback by applying sentiment analysis to improve the quality of teaching and learning in the education sector have yielded the required results. Given the above this research effort aimed at reviewing published papers geared towards improving learners' experiences and enhancing quality assurance in the education system from published articles between 2016 to 2023 in referenced databases from high-impact journals. The research study collated 80 useful articles from the pool of 4595 that were earlier surfed using eligibility criteria which entailed employing exclusion and inclusion characteristics. The outcome of the literature reviews showed that 70% of researchers have used sentiment analysis as feedback and improvement analysis tools in enhancing the student learning experience and quality assurance in the higher education sector. The highly used sentiment analysis algorithms are lexicon-based, Support Vector Machine (SVM), Naïve Bayes(NB), Linear Regression(LR), and Artificial Neural Network(ANN). Keywords: Sentiment Analysis, Education Sector, Supervised Machine Learning Algorithms, social media, Questionnaires, Lexicon-based, Natural Language Processing(NLP) Emughedi Oghu, Emeka Ogbuju, Taiwo Abiodun & Francisca Oladipo.. (2023): Baseline Study of Different Sentiment Analysis Computing Methods to Enhance Quality Assurance In Teaching and Learning. Journal of Advances in Mathematical & Computational Science. Vol. 10, No. 4. Pp 1-20 Available online at www.isteams.net/mathematics-computationaljournal. dx.doi.org/10.22624/AIMS/MATHS/V11N3P1