A New Paradigm Through Machine Learning: A Learning Maximization Approach for Sustainable Education

Abstract

Machine learning algorithms have been used to predict student performance in academic institutions over the last decade. The developed prediction models classified students as either those who were likely to receive a distinctive or those who were likely to be at-risk or withdraw from the class. Although, the early prediction aid in the targeting of educational interventions and achieved a more efficient allocation of educational resources. All proposed solutions fell within the scope of predictions that result in active or proactive actions to support universities and learners. On the other hand, they fail to comprehend the various forms of education systems and whether it appropriate for the twenty-first century and future generations. The paper classifies education into five types based on the design mode, the scope of production, and the interaction between learners and educational systems (Push, Pull, coupling, Integrated, and Sustainable). The paper proposes a sustainable education paradigm that maximizes the knowledge and skill matrix accumulated for the desired program. The proposed theory implementation phases are modelled and demonstrated using 21st-century technologies, such as personalized and coaching education based on the learner’s learning style and remediation actions for strong learners with innovative competencies. The study emphasized various aspects of sustainable education systems that are required for smart city transition. The limitations and proposed solutions for dealing with anticipated issues were demonstrated, and the benefits of sustainable education are based on the proposed maximization theory rather than the current block-based learning of outcomes.