The goal of this study is to synthesize the findings, methodology and research themes of peer reviewed studies on Artificial Intelligence in higher education, published between 2000 to 2020. Twenty-nine articles were selected for review by following the PRISMA approach. The demographical and thematic trends suggest that most research is skewed towards few geographical locations (USA, Europe, India, China, Hong-Kong) and recent time periods (2018-2020) and scattered across publications from varied disciplinary traditions. Taiwan and United States contributed most to the number of studies, with 2017 being the most fruitful year. Vectors as well as decision trees were the most often used machine learning algorithms. Mechanization, cognitive process assessment, prediction models, integrated learning systems, and tackling potential problems in the use of big data and learning analytics were among the most commonly explored topics. Expanding geographical variety, adopting advanced algorithmic approaches including Bayesian as well as fuzzy logic techniques in educational machine learning work; applications for knowledge-based systems, and personalized learning were suggested for future search. Conclusions are drawn and future research directions identified. Potential research recommendations emphasize the expansion of geographical, topical, and methodological variety.
Gera, R., & Chadha, P. (2021). Systematic review of artificial intelligence in higher education (2000-2020) and future research directions. In W. B. James, C. Cobanoglu, & M. Cavusoglu (Eds.), Advances in global education and research (Vol. 4, pp. 1–12). USF M3 Publishing. https://www.doi.org/10.5038/9781955833042
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License