Evaluation of learners performance using computational intelligence and machine learning algorithms

Abstract

Educational Data Analytics (EDA) is a multidisciplinary field newlineutilizing data mining, statistical analysis, and machine learning to extract newlinevaluable insights from diverse educational datasets. This research work reviews newlineexisting literature in EDA, highlighting research contributions that leverage newlinemachine learning algorithms for predicting and enhancing student performance. newlineVarious studies employ innovative methods, including particle swarm newlineoptimization and novel scalable algorithms, showcasing impressive accuracy newlinerates in assessing student behavior and performance. newlineThe research identifies a gap in addressing the holistic assessment newlineof Knowledge, Skills, and Attitude (KSA) towards academic performance newlinein real-time. To bridge this gap, new approaches are proposed to employ newlinecomputational intelligence and machine learning algorithms. The research newlineobjectives include evaluating the impact of Learning Management Systems newline(LMS) and Information and Communication Technology (ICT) tools on newlineacademic performance, developing machine learning based systems for newlineperformance prediction, and implementing hybrid algorithms combining newlinefuzzy logic and machine learning algorithm. newlineThe proposed work encompasses a self-report questionnaire to collect newlinelearner responses, statistical measures to evaluate academic performance, and newlinethe development of metaheuristic algorithms such as Chaotic Particle Swarm newlineOptimization (C-PSO). Results indicate improvements in KSA and academic newlineperformance, with C-PSO outperforming other metaheuristic algorithms. newline newline