Abstract
Smart education requires design, development, implementation and active use of innovative systems, technologies, teaching and learning strategies and approaches that are based on various data sources in academia, modern mathematical methods in data statistics and data analytics, and state-of-the-art data-driven approaches and technologies. The availability of tools that measure, collect, clean, organize, analyze, process, store, visualize and report data about student academic performance in an academic course and/or student overall academic progress in the selected program of study has given rise to the field of learning analytics for student academic success. Student data representation, processing and prediction, as a central part of learning analytics system, are crucial topics for researchers and practitioners in academia. Our vision for the engineering of smart learning analytics—the next generation of learning analytics—is based on the concept that this technology should strongly support (a) various “smartness” levels of smart education such as adaptivity, sensing, inferring, anticipation, self-learning and self-organization, and (b) main types of data analytics of smart education such as descriptive, diagnostic, predictive and prescriptive analytics. This paper presents the up-to-date findings and outcomes of the research, design and development project at the InterLabs Research Institute at Bradley University (USA) aimed at application of a quantitative approach to student academic performance data representation, hierarchical levels of data processing, multiple quality evaluation criteria to be selected and used, and high-quality student academic performance data prediction in smart learning analytics systems.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call