Improving Learning Outcomes through Predictive Analytics: Enhancing Teaching and Learning with Educational Data Mining
Ashraf Alam
Abstract
Educational Data Mining (EDM) is a promising area of research that leverages computational methods to improve educational outcomes by extracting valuable insights from vast educational datasets. The purpose of this study is to explore the relationship between EDM and various educational theories, including Learning Analytics Theory (LAT), Educational Psychology Theory, Cognitive Load Theory, Self-Regulated Learning (SRL) Theory, Bloom’s Taxonomy Theory, Multiple Intelligences Theory, Schema Theory, Situated Learning Theory, Zone of Proximal Development Theory, and Connectivism Theory. This scientific research provides a comprehensive overview of each theory, and discusses how EDM can be used to enhance the understanding and application of these theories in educational settings. Through various teaching-learning classroom examples, the study illustrates how EDM can help identify students’ learning styles, strengths, and weaknesses, develop algorithms that adapt to students’ learning needs in real-time, predict students’ future academic performance, identify challenging areas of a course, and provide tailored instruction and support to individual students. The study further demonstrates how EDM can inform the design of effective teaching strategies, and contribute to the development of personalized and adaptive learning environments.