Evaluating students’ academic performance using clustering techniques (a case study of school of engineering, federal polytechnic, bida

Abstract

Predicting students’ performance becomes more challenging due to large volume of data in educational databases. Data Clustering is the task of grouping a set of objects in such a way that objects in the same group are more similar to each other than to those in other groups. Clustering categorizes data into groups such that objects are grouped in the same cluster when they are similar according to specific metrics. It is one of the methods in data mining to analyse the massive volume of data. With clustering, interesting patterns and structures can be found directly from very large data sets with little or no background knowledge. This work evaluates students’ academic performance using clustering techniques. Data on examination raw scores of final year Higher National Diploma (HND) students from five departments of School of Engineering, The Federal Polytechnic, Bida, were collected from the Examination and Records Unit of the Polytechnic. For each department, three clustering algorithms (namely, the kMeans, k-Medoids, and Fuzzy C Means (FCM)) were used to organize the collected data (the students’ examination raw scores) into three clusters based on similarity. These clusters define the students’ academic ability (performance) and they include average, good and excellent performance. The clustering algorithms were then compared based on Cluster Size, Cluster membership, and number of iterations taken to reach convergence. Recommendations for further studies were made at the end of the work.