Authors: Mr. Nikhil Barapatre, Ms. Ruchika Kadu, Mr. Uday Mahure, Ms. Vaishnavi Nawale
Abstract: The rapid advancement of digital technologies has significantly transformed the education sector by introducing intelligent learning environments, online learning platforms, Learning Management Systems (LMS), Massive Open Online Courses (MOOCs), and virtual classrooms. These technological developments generate enormous volumes of educational data that capture students' learning activities, academic performance, attendance, assignment submissions, examination scores, participation in online discussions, and interaction with digital learning resources. The availability of such educational data has created new opportunities for educational institutions to improve teaching strategies, personalize learning experiences, and enhance academic outcomes through data-driven decision-making. Learning Analytics has emerged as a multidisciplinary research field that combines Artificial Intelligence (AI), Educational Data Mining (EDM), Machine Learning (ML), Data Analytics, and statistical techniques to analyze educational data and predict student performance with greater accuracy. By identifying learning patterns and behavioral characteristics, learning analytics enables educators to provide timely interventions that support student success and reduce academic failure. Student performance prediction has become one of the most important applications of learning analytics because academic achievement directly influences educational quality, institutional effectiveness, and workforce development. Traditional evaluation methods primarily depend on periodic examinations and manual assessment, which often identify learning difficulties only after students experience poor academic performance. Such delayed identification limits the opportunity for educators to provide timely academic support. In contrast, learning analytics continuously monitors students' learning behavior throughout the educational process, enabling early detection of students who may be at risk of poor performance, course failure, or dropout. This proactive approach allows instructors to implement personalized learning strategies and academic interventions before significant learning problems occur. This research investigates the application of learning analytics for predicting student academic performance using multiple educational indicators collected from both traditional and digital learning environments. The proposed study considers a comprehensive set of predictive variables, including attendance records, assignment completion rates, quiz performance, laboratory activities, examination marks, participation in online learning platforms, classroom engagement, discussion forum activities, learning duration, assessment consistency, and demographic information.
DOI: http://doi.org/10.5281/zenodo.21486978
