Continuous Clustering in Big Data Learning Analytics

Abstract

Learners' attainment of academic knowledge in postsecondary institutions is predominantly expressed by summative or formative assessment approaches. Recent advances in educational technology has hinted at a means to measure learning efficiency, in terms of personalization of learner competency and capacity in terms of adaptability of observed practices, using raw data observed from study experiences of learners as individuals and as contributors in social networks. While accurate computational models that embody learning efficiency remain a distant and elusive goal, big data learning analytics approaches this goal by recognizing competency growth of learners, at various levels of granularity, using a combination of continuous, formative and summative assessments. This study discusses a method to continuously capture data from students' learning interactions. Then, it analyzes and clusters the data based on their individual performances in terms of accuracy, efficiency and quality by employing Particle Swarm Optimization (PSO) algorithm.