Service Quality in Higher Education: A Bibliometric Visualization and Prediction of Research Productivity and Thematic Evolution Using Biblioshiny and BERTopic

Abstract

This study maps, models, and forecasts the evolution of service quality research in higher education using 951 articles indexed in Scopus and Web of Science (1993–2024). Biblioshiny was employed to analyze research productivity, collaboration patterns, influential authors, and keyword evolution, while BERTopic and Dynamic Topic Modeling (DTM) identified and tracked thematic developments over time. Topic interrelationships were examined through chord diagrams, and Prophet forecasting was used to predict topic prevalence through 2028. The findings reveal Malaysia as the most productive country and the UK and USA as leading collaborators, with Parvez Sultan emerging as the most prolific author. Keyword trends indicate a shift from quality assurance and student satisfaction toward digital learning, e-learning, and technology-enabled education. BERTopic identified 20 latent themes, with service_quality_customer as the dominant topic and learn_system_learning_student as the fastest-growing theme. Forecasts suggest continued expansion of student-centered, learning-focused, and technology-driven research, with the field expected to reach its publication peak around 2042. The study provides practical insights for higher education institutions and policymakers by highlighting the growing importance of student satisfaction, teaching quality, and digital learning infrastructure. By integrating bibliometric analysis, topic modeling, thematic evolution, and forecasting, this study offers a comprehensive and predictive understanding of service quality research in higher education.