Analysing the Academic Performance of Students Using Unsupervised Data Mining

Abstract

Educational Data Mining is an attractive interdisciplinary domain in which the main goal is to apply data mining techniques in educational environments in order to offer better insights into the educational related tasks. This paper analyses the relevance of two unsupervised learning models, self-organizing maps and relational association rule mining in the context of students' performance prediction. The experimental results obtained by applying the aforementioned unsupervised learning models on a real data set collected from Babeş-Bolyai University emphasize their effectiveness in mining relevant relationships and rules from educational data which may be useful for predicting the academic performance of students.

IntelliDaM: A Machine Learning-Based Framework for Enhancing the Performance of Decision-Making Processes. A Case Study for Educational Data Mining

“…Recent approaches in SPA addressed unsupervised RAR mining and SOMs [49], [50], to extract from academic data sets patterns and rules relevant for analysing students' academic performance. Other studies compare the ability of autoencoders and SOMs to find learning patterns in data sets related to students' performance in traditional and online environments [51], [52].…”