Ten Essential Pillars in Artificial Intelligence for University Science Education: A Scoping Review

Abstract

Although Artificial Intelligence (AI) is notable in education, the studies on its specific application in university science education are still incipient. At the same time, the research demonstrates a critical need to systematize AI pillars to provide a coherent and clear structure to guide the development, implementation, and understanding of this technology in various fields, but very little progress has been made in the field of university science education. Therefore, the present study was aimed at exploring the essential pillars of AI for university science education. This scoping review followed the Arksey and O’Malley methodology, which unfolds five stages; based on established criteria 89 texts were finally selected and included in the study. Ten pillars were found: (1) AI Ethics, (2) AI Didactic Integration (AI-DI), (3) Machine Learning (ML), (4) Deep Learning (DL), (5) Active Learning (AL), (6) Intelligent Prediction (AI-IP), (7) Natural Language Processing (NLP), (8) Augmented reality and Virtual reality (AR/VR), (9) Artificial Neural Network (ANN), and (10) Intelligent Tutoring System (ITS). The study provides a comprehensive synthesis of current trends and advances in this field, highlighting good practices that provide empirical evidence, highlighting ethical, pedagogical, and technical challenges associated with the application of AI in science education, which can contribute to the formation of an aware and ethical educational community in the use of AI.