Clustering Student Data to Characterize Performance Patterns

Abstract

Over the years the academic records of thousands of students have accumulated in educational institutions and most of these data are available in digital format. Mining these huge volumes of data may gain a deeper insight and can throw some light on planning pedagogical approaches and strategies in the future. We propose to formulate this problem as a data mining task and use k-means clustering and fuzzy c-means clustering algorithms to evolve hidden patterns.